CN121365166A - Data set recommendation method and device, electronic equipment and readable storage medium - Google Patents

Abstract

This application relates to a method, apparatus, electronic device, computer-readable storage medium, and computer program product for recommending data collections. The method includes: acquiring target tags; aggregating target data matching the target tags into a corresponding recommendation collection; the target data being recorded from at least two applications; the target data being obtained based on preset recording operations; and displaying the recommendation collection. This method enables automatic and accurate recommendation of data collections.

Description

Data set recommendation method and device, electronic equipment and readable storage medium

Technical Field

The present application relates to the field of information technology, and in particular, to a data set recommendation method, apparatus, electronic device, computer readable storage medium, and computer program product.

Background

With the rapid development of computer and network technologies, electronic devices are increasingly used, have more functions, and become an indispensable part of daily life. The user can record the drops in work and life through the electronic equipment and store the drops in the form of one piece of data, however, under the condition of more stored data, the data content really needed by the user cannot be presented through simple searching, so that a strategy capable of automatically presenting the data set matched with the real requirements of the user is needed.

Disclosure of Invention

The embodiment of the application provides a data set recommendation method, a device, electronic equipment and a computer readable storage medium, which can automatically recommend a data set matched with the real requirement of a user to the user.

In a first aspect, the present application provides a data set recommendation method, including:

obtaining a target label;

aggregating target data matched with the target tag into a corresponding recommended collection, wherein the target data is recorded from at least two application programs, and the target data is obtained based on a preset recording operation;

And displaying the recommendation set.

In a second aspect, the present application further provides a data set recommendation device, including:

the label acquisition module is used for acquiring the target label;

the data aggregation module is used for aggregating target data matched with the target tag into corresponding recommendation aggregate, wherein the target data is recorded from at least two application programs, and the target data is obtained based on a preset recording operation;

and the collection display module is used for displaying the recommended collection.

In a third aspect, the present application also provides an electronic device, comprising a memory storing a computer program and a processor implementing the steps of the data set recommendation method provided in the first aspect when the computer program is executed by the processor.

In a fourth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the data set recommendation method provided in the first aspect.

In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the data set recommendation method provided in the first aspect.

According to the data collection recommendation method, the device, the electronic equipment, the computer readable storage medium and the computer program product, the data matched with the target label is selected and aggregated into the collection and then recommended to the user by acquiring the target label and then aggregating the target data matched with the target label into the corresponding recommendation collection and displaying the recommendation collection, so that the accuracy of recommendation is improved, meanwhile, the data information matched with the real requirements of the user is automatically provided for the user, and the use experience of the equipment can be improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are needed in the description of the embodiments of the present application or the related technologies will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other related drawings may be obtained according to these drawings without inventive effort to those of ordinary skill in the art.

FIG. 1 is a block diagram of a computer system executing a visual or multimodal search service in some embodiments;

FIG. 2 is a schematic diagram of a frame of a data processing system of an electronic device in some embodiments;

FIG. 3 is a schematic diagram of the internal architecture of a data processing system of an electronic device in some embodiments;

FIG. 4 is a flow chart of a data collection recommendation method in some embodiments;

FIG. 5 is a schematic diagram of a display of a recommendation set in some embodiments;

FIG. 6 is a schematic diagram of a record data display in a target recommendation set in some embodiments;

FIG. 7 is a schematic diagram of a display after a "keep-fit" control is triggered in some embodiments;

FIG. 8 is a schematic diagram of a display after the "no interest" control is triggered in some embodiments;

FIG. 9 is a schematic display of a candidate functionality control in some embodiments;

FIG. 10 is a flow chart of a data collection recommendation method in other embodiments;

FIG. 11 is a block diagram of a data collection recommendation device in some embodiments.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The data set recommendation method provided by the embodiment of the application can be applied to electronic equipment. The electronic device may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, aircrafts, unmanned aerial vehicles, internet of things devices and portable wearable devices, and the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, projection devices and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The head-mounted device may be a Virtual Reality (VR) device, an augmented Reality (Augmented Reality, AR) device, smart glasses, or the like. The server 104 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud computing services. The electronic device may be a terminal or a server.

FIG. 1 depicts a block diagram of an example computing system 100 executing a visual or multimodal search service, according to an example embodiment of the application. Computing system 100 includes a user computing system 110, a server computing system 130, and/or a third party computing system 150 communicatively coupled by a network 160.

The user computing system 110 may include any type of computing device, such as, for example, a personal computing device (e.g., a laptop computer or desktop computer), a mobile computing device (e.g., a smart phone or tablet computer), a game console or controller, a wearable computing device (e.g., a smart watch or smart glasses, etc.), an embedded computing device, or any other type of computing device.

The user computing system 110 includes one or more processors 112 and memory 114. The one or more processors 112 may be any suitable processing device (e.g., a processor core, microprocessor, ASIC (Application SPECIFIC INTEGRATED Circuit), FPGA (Field Programmable GATE ARRAY ), controller, microcontroller, etc.) and may be one processor or multiple processors operatively connected. Memory 114 may include one or more non-transitory computer-readable storage media such as RAM (Random Access Memory ), ROM (Read-Only Memory), EEPROM (ELECTRICALLY ERASABLE PROGRAMMABLE READ-Only Memory ), EPROM (Erasable Programmable Read-Only Memory), flash Memory devices, magnetic disks, and the like, as well as combinations thereof. Memory 114 may store data 116 and instructions 118 that are executed by processor 112 to cause user computing system 110 to perform operations.

In some implementations, the user computing system 110 may store or include one or more machine learning models 120. For example, the machine learning model 120 may be or otherwise include various machine learning models, such as a neural network (e.g., a deep neural network) or other types of machine learning models, including nonlinear models and/or linear models. The neural network may include a feed forward neural network, a recurrent neural network (e.g., a long and short term memory recurrent neural network), a convolutional neural network, or other form of neural network.

In some implementations, one or more machine learning models 120 may be received from the server computing system 130 over the network 160, stored in the memory 114 of the user computing system 110, and then used or otherwise implemented by the one or more processors 112. In some implementations, the user computing system 110 may implement multiple parallel instances of a single machine learning model 120 (e.g., to perform parallel machine learning model processing across multiple instances of input data and/or detected features).

Additionally or alternatively, one or more machine learning models 140 may be included in the server computing system 130 or otherwise stored and implemented by the server computing system 130, the server computing system 130 in communication with the user computing system 110 according to a client/server relationship. For example, the machine learning model 140 may be implemented by the server computing system 130 as part of a web service (e.g., viewfinder service, visual search service, image processing service, environmental computing service, and/or overlay application service). Accordingly, one or more machine learning models 120 may be stored and implemented at the user computing system 110 and/or one or more machine learning models 140 may be stored and implemented at the server computing system 130.

The machine learning model 120 or 140 may include one or more generative models, one or more object detection models, one or more segmentation models, one or more classification models, one or more embedding models, one or more semantic analysis models, and/or one or more search engines, among others.

The one or more generation models may be used to process the display data and/or the one or more processing outputs to generate natural language output (e.g., natural language output including additional information about the display data and/or entities associated with data depicted in the displayed content), generate images, and/or other model-generated media content items. For example, one or more web resources may be accessed and processed to generate a summary of a particular topic. One or more object detection models may be used to perform object detection in the display data. One or more segmentation models may be used to segment objects and/or text segments from the displayed content. One or more classification models may be used to perform object classification, image classification, entity classification, format classification, emotion classification, and/or other classification tasks. One or more embedding models may be used to embed portions and/or all of the display data. The embedding may then be utilized to search for similar objects and/or text, classification, grouping, and/or compression. The semantic analysis model may be used to process the display data to generate semantic outputs that describe understanding of the display data with respect to topic understanding, scene understanding, focus, pattern recognition, application understanding, and/or one or more other semantic outputs.

One or more search engines may process the display data, portions of the display data, and/or one or more machine learning model outputs to determine one or more search results. The one or more search results may include web pages, images, text, video, and/or other data. The search results may be determined based on feature mapping, feature matching, embedded searching, metadata searching, tag searching, clustering, and/or other search techniques. The search results may be determined based on query intent classification, search result classification, and/or entity classification. The output of the model and/or the search results may be sent back to the user computing device for provision to the user through one or more user interface elements generated and provided by the visual search interface.

The user computing system 110 may also include one or more user input components 122 that receive user input. For example, the user input component 122 may be a touch-sensitive component (e.g., a touch-sensitive display screen or touchpad) that is sensitive to touch by a user input object (e.g., a finger or stylus). The touch sensitive component may be used to implement a virtual keyboard. Other example user input components include a microphone, a conventional keyboard, or other device by which a user may provide user input.

In some implementations, the user computing system 110 can store and/or provide one or more user interfaces 124, which user interfaces 124 can be associated with one or more applications. The one or more user interfaces 124 may be configured to receive input and/or provide data for display (e.g., image data, text data, audio data, one or more user interface elements, an augmented reality experience, a virtual reality experience, and/or other data for display). The user interface 124 may be associated with one or more other computing systems (e.g., the server computing system 130 and/or the third party computing system 150). User interface 124 may include a viewfinder interface, a search interface, a generative model interface, a social media interface, a media content gallery interface, and the like.

The user computing system 110 may include one or more sensors 126 and/or receive data from one or more sensors 126. The one or more sensors 126 may be housed in a housing component that houses the one or more processors 112, the memory 114, and/or one or more hardware components that may store and/or cause execution of one or more software packages. The one or more sensors 126 may include one or more image sensors (e.g., cameras), one or more radar sensors, one or more audio sensors (e.g., microphones), one or more inertial sensors (e.g., inertial measurement units), one or more biological sensors (e.g., heart rate sensors, pulse sensors, retinal sensors, and/or fingerprint sensors), one or more infrared sensors, one or more location sensors (e.g., global positioning system, GPS), one or more touch sensors (e.g., conductive touch sensors and/or mechanical touch sensors), and/or one or more other sensors. One or more sensors may be utilized to obtain data associated with the user environment (e.g., an image of the user environment, a record of the environment, and/or a location of the user).

The user computing system 110 may include a user computing device 111 and/or be part of the user computing device 111. User computing device 111 may include a mobile computing device (e.g., a smart phone or tablet computer), a desktop computer, a laptop computer, a smart wearable device, and/or a smart appliance or aircraft or vehicle-mounted device, among others. Additionally and/or alternatively, the user computing system 110 may obtain data from one or more user computing devices 111 and/or generate data using one or more user computing devices 111. For example, image data depicting the environment may be captured with a camera of a smartphone, and/or data provided to the user may be tracked and/or processed with an overlay application of the user computing device 111. Similarly, data about the user and/or about the user's environment may be obtained using one or more sensors associated with the smart wearable device (e.g., image data may be obtained using a camera housed in the user's smart glasses). Additionally and/or alternatively, data may be obtained and uploaded from other user devices that may be dedicated to data acquisition or generation.

The server computing system 130 includes one or more processors 132 and memory 134. The one or more processors 132 may be any suitable processing device (e.g., a processor core, microprocessor, ASIC, FPGA, controller, microcontroller, etc.) and may be one processor or operatively connected multiple processors. Memory 134 may include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, flash memory devices, disks, and the like, as well as combinations thereof. Memory 134 may store data 136 and instructions 138 that are executed by processor 132 to cause server computing system 130 to perform operations.

In some implementations, the server computing system 130 includes or is otherwise implemented by one or more server computing devices. Where the server computing system 130 includes multiple server computing devices, such server computing devices may operate in accordance with a sequential computing architecture, a parallel computing architecture, or some combination thereof.

As described above, the server computing system 130 may store or otherwise include one or more machine learning models 140. For example, the machine learning model 140 may be or may otherwise include various machine learning models. Example machine learning models include neural networks or other multi-layer nonlinear models. Example neural networks include feed forward neural networks, deep neural networks, recurrent neural networks, and convolutional neural networks.

Additionally and/or alternatively, the server computing system 130 can include a search engine 142 and/or be communicatively connected with the search engine 142, which can be used to crawl one or more databases (and/or resources). Search engine 142 may process data from user computing system 110, server computing system 130, and/or third party computing system 150 to determine one or more search results associated with the input data. Search engine 142 may perform term-based searches, tag-based searches, boolean searches, image searches, embedded-based searches (e.g., nearest neighbor searches), multimodal searches, and/or one or more other search techniques.

The server computing system 130 may store and/or provide one or more user interfaces 144 for obtaining input data and/or providing output data to one or more users. The one or more user interfaces 144 may include one or more user interface elements that may include input fields, navigation tools, content tiles, selectable tiles, widgets, data display carousel, dynamic animations, information popup windows, image enhancements, text-to-speech, speech-to-text, augmented reality, virtual reality, feedback loops, and/or other interface elements.

The user computing system 110 and/or the server computing system 130 may train the machine learning model 120 and/or 140 via interaction with a third party computing system 150 communicatively coupled via a network 160. The third party computing system 150 may be separate from the server computing system 130 or may be part of the server computing system 130. Alternatively and/or additionally, the third party computing system 150 may be associated with one or more network resources, one or more network platforms, one or more other users, and/or one or more contexts.

The third party computing system 150 may include one or more processors 152 and memory 154. The one or more processors 152 may be any suitable processing device (e.g., processor core, microprocessor, ASIC, FPGA, controller, micro-processor)

Controller, etc.) and may be one processor or a plurality of processors operatively connected. Memory 154 may include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, flash memory devices, disks, and the like, as well as combinations thereof. The memory 154 may store data 156 and instructions 158 that are executed by the processor 152 to cause the third party computing system 150 to perform operations. In some implementations, the third party computing system 150 includes or is otherwise implemented by one or more server computing devices.

The network 160 may be any type of communication network, such as a local area network (e.g., an intranet), a wide area network (e.g., the internet), or some combination thereof, and may include any number of wired or wireless links. In general, communication over the network 180 may be performed via any type of wired and/or wireless connection using a variety of communication protocols (e.g., TCP/IP, HTTP, SMTP, FTP), coding or formats (e.g., HTML, XML), and/or protection schemes (e.g., VPN, secure HTTP, SSL).

The machine learning model described in the present disclosure may be used for various tasks, applications, and/or use cases.

In some implementations, the input to the machine learning model of the present application can be image data. The machine learning model may process the image data to generate an output. As one example, the machine learning model may process the image data to generate an image recognition output (e.g., recognition of the image data, potential embedding of the image data, encoded representation of the image data, hash of the image data, etc.). As another example, the machine learning model may process the image data to generate an image segmentation output. As another example, the machine learning model may process image data to generate an image classification output. As another example, the machine learning model may process the image data to generate an image data modification output (e.g., a change in the image data, etc.). As another example, the machine learning model may process the image data to generate an encoded image data output (e.g., an encoded and/or compressed representation of the image data, etc.). As another example, the machine learning model may process image data to generate an upgraded image data output. As another example, the machine learning model may process image data to generate a prediction output.

In some implementations, the input of the machine learning model of the present disclosure can be text or natural language data. The machine learning model may process text or natural language data to generate an output. As one example, the machine learning model may process natural language data to generate a language encoded output. As another example, the machine learning model may process text or natural language data to generate a potential text-embedded output. As another example, the machine learning model may process text or natural language data to generate a translation output. As another example, the machine learning model may process text or natural language data to generate a classification output. As another example, the machine learning model may process text or natural language data to generate a text segmentation output. As another example, the machine learning model may process text or natural language data to generate semantic intent output. As another example, the machine learning model may process text or natural language data to generate an upgraded text or natural language output (e.g., text or natural language data of higher quality than the input text or natural language, etc.). As another example, the machine learning model may process text or natural language data to generate a predictive output.

In some implementations, the input to the machine learning model of the present disclosure can be speech data. The machine learning model may process the speech data to generate an output. As one example, the machine learning model may process speech data to generate a speech recognition output. As another example, the machine learning model may process speech data to generate speech translation output. As another example, the machine learning model may process speech data to generate potential embedded outputs. As another example, the machine learning model may process the speech data to generate an encoded speech output (e.g., an encoded and/or compressed representation of the speech data, etc.). As another example, the machine learning model may process the speech data to generate an upgraded speech output (e.g., speech data of higher quality than the input speech data, etc.). As another example, the machine learning model may process the speech data to generate a textual representation output (e.g., a textual representation of the input speech data, etc.). As another example, the machine learning model may process speech data to generate a prediction output.

In some implementations, the input to the machine learning model of the present disclosure can be sensor data. The machine learning model may process the sensor data to generate an output. As one example, the machine learning model may process the sensor data to generate an identification output. As another example, the machine learning model may process the sensor data to generate a prediction output. As another example, the machine learning model may process the sensor data to generate a classification output. As another example, the machine learning model may process the sensor data to generate a segmented output. As another example, the machine learning model may process the sensor data to generate a segmented output. As another example, the machine learning model may process the sensor data to generate a visual output. As another example, the machine learning model may process the sensor data to generate a diagnostic output. As another example, the machine learning model may process the sensor data to generate a detection output.

In some cases, the input includes visual data and the task is a computer visual task. In some cases, pixel data including one or more images is input, and the task is an image processing task. For example, an image processing task may be an image classification, wherein the output is a set of scores, each score corresponding to a different object class and representing a likelihood that one or more images depict an object belonging to the object class.

The user computing system 110 may include a plurality of applications (e.g., applications 1 through N). Each application may include its own respective machine learning library and machine learning model. For example, each application may include a machine learning model. Example applications include text messaging applications, email applications, dictation applications, virtual keyboard applications, browser applications, and the like. In some implementations, each application may use an API (e.g., a public API across all applications) to communicate with the central intelligence layer (and the models stored therein).

The central intelligence layer may include a plurality of machine learning models. For example, a respective machine learning model (e.g., model) may be provided for each application, and managed by a central intelligent layer. In other implementations, two or more applications may share a single machine learning model. For example, in some implementations, the central intelligence layer can provide a single model (e.g., a single model) for all applications. In some implementations, the central intelligence layer is included within or otherwise implemented by an operating system of the computing system 1100.

The central intelligence layer may communicate with the central device data layer. The central device data layer may be a centralized data store of the computing system 100. The central device data layer may communicate with many other components of the computing device, such as, for example, one or more sensors, a context manager, a device status component, and/or additional components. In some implementations, the central device data layer may communicate with each device component using an API (e.g., a proprietary API).

The user computing device 111 may be an electronic device. FIG. 2 is a schematic diagram of a data processing system of an electronic device 200 in some embodiments. The electronic device 200 includes a data acquisition module 210, a large model 220, and a database 230. The electronic device 200 can implement a series of functions of data collection, preprocessing, storage, mining, retrieval, recommendation, question-answering, and the like for various applications.

In the process of displaying the application page content or the web page content, the electronic device 200 may detect that the user starts the data recording function through a preset triggering manner, where the preset triggering manner may be triggering a target function key, gesture triggering (such as sliding upwards in three directions), or triggering of the data recording application program.

The data collection module 210 is configured to record, according to the opened data recording function, screen display content of the electronic device 200 or page content of an application running in the background. The screen display content or page content of the background running application may include data content such as text, pictures, video, audio, and the like. If the text is the text, the text data can be directly acquired. If the picture is the picture, the picture can be downloaded according to the address of the picture. If the video is the video, the URL address of the video can be obtained, and all or fragment contents of the video can be downloaded according to the URL address. If the audio is the audio, the download address of the audio can be obtained, and all or part of the data of the audio can be downloaded. The data acquisition module 210 may also record the screen display content in a screenshot mode, or acquire voice data input by a user through a microphone for preprocessing.

The data collection module 210 is further configured to invoke the large model 220 or some algorithms to preprocess the recorded screen display content, so as to obtain data in a target format. For example, if the screen display content is text, entity extraction and summary can be performed, if the screen display content is a picture, the picture can be identified to obtain picture content and classify the picture, if the screen display content is video, key frame extraction and key frame content identification can be performed on the video, and summary information can be generated on the video. The processed data may be stored in database 230. Wherein the large model 220 may be the machine learning model of fig. 1.

Database 230 is a carrier or system for storing, managing and retrieving information in a form that may vary depending on the object of storage, the context of use and the architecture of technology. Database 230 may store data and may have at least one of classification, retrieval, synchronization, rights control, and the like. The database 230 may be an application, applet, API interface, personal specific storage hardware, embedded hardware logging module, cloud service form database, browser plug-in, bookmark management tool, enterprise knowledge base system, cache type database, personal information base, etc. The personal special storage hardware can be a mobile hard disk with a special management system, an encrypted U disk, a personal cloud storage hard disk and the like. The embedded hardware recording module can be a special storage module in the intelligent equipment and is used for recording various data. The database of the cloud service form can be a cloud database providing interface calling service functions or a database of personal cloud data management service.

Database 230 includes a data management module 240 and a data interaction module 250. The data management module 240 may invoke the large model 220 to mine the stored data and store the mined data in the database 230, may provide data retrieval functionality, and the like. Mining modes can comprise data desensitization processing, label classification, entity extraction, schedule extraction, to-do item extraction, abstract generation, generation of a collection, recommendation of a collection, data association and the like. The data management module 240 may further perform induction and sorting according to various data recorded by the user, so as to obtain user portrait information. The user portrait information may include user personal information, personal preference information, and the like. The user personal information may include information of identity information, occupation, academy, family, etc. of the user, and the personal preference information may include favorite people, things, places, scenery, games, etc., without being limited thereto.

The data interaction module 250 may provide an interaction function with an application program, and may acquire data required for interaction through the data management module 240. The interactive functions may include, but are not limited to, recommendation collection, data search, data detail viewing, data sharing, AI questions and answers, data association, and the like. For example, the AI question and answer function, the data interaction module 250 acquires a question provided by the user in the AI question and answer application and transmits the question to the data management module 240, and the data management module 240 searches for a corresponding answer and provides the answer to the data interaction module 250 for feedback to the AI question and answer related application.

Further, the data collected by the data collection module 210 and the mined user portrait information in the electronic device 200 can be synchronized to the cloud end according to the needs, the collected data can be processed by means of a large cloud end model, and the data synchronized to the cloud end can be synchronized to other electronic devices through the cloud end, so that sharing of the data among different devices is realized.

FIG. 3 is a schematic diagram of the internal architecture of a data processing system of an electronic device in some embodiments. The data acquisition module 210 may include an acquisition entry sub-module 211, an extraction sub-module 212, a screenshot acquisition sub-module 213, a preprocessing sub-module 214, a writing sub-module 215, and the like.

The collection entry sub-module 211 is configured to provide an entry for data collection, for example, a voice assistant entry, through which a data recording instruction of a user voice input is received, or a gesture trigger entry, such as a long entry of a hardware key, or a three-way sliding entry, or a trigger entry of a user for a specific control, such as a praise control, a collection control, a sharing control, or the like. The acquisition inlet sub-module 21 can automatically acquire the page information according to the set rule, if the acquisition triggering condition is the preset page content, and if the preset page content exists in the page information displayed by the electronic equipment, the page information is automatically acquired. The collection inlet sub-module 21 may also receive donation data of some application programs (such as weather application, clock application, step counting application, etc.), such as weather data, time data, walking data, running mileage, etc., but is not limited thereto.

The extraction sub-module 212 may perform image-text extraction on the page content of the application to be recorded.

The screenshot obtaining sub-module 213 may obtain text content by screenshot the screen content and then invoking the screenshot content of the OCR recognition algorithm to perform text recognition.

The preprocessing sub-module 214 may segment or denoise or process the collected data into data in a preset format.

The writing sub-module 215 is used to write the collected data or the preprocessed data into the database 230.

The data management module 240 may include a data mining sub-module 241, a question and answer sub-module 242, a retrieval sub-module 243, a source data processing sub-module 244, and the like.

The data mining sub-module 241 is used for mining data. Mining modes can comprise data desensitization processing, label classification, entity extraction, schedule extraction, to-do item extraction, abstract generation, generation of a collection, recommendation of a collection, data association and the like.

The question and answer sub-module 242 is used for searching corresponding data according to the user questions and generating corresponding answers according to the searched data.

The retrieving sub-module 243 is configured to retrieve corresponding data from the database 230 according to the search request, and sort or directly feed back the retrieved data to the user.

The source data processing sub-module 244 is configured to provide add, modify, delete, update operations on the source data. The source data may refer to data recorded in the database 230 after collection or data after preprocessing the collected data.

The data interaction module 250 includes a data addition/editing/viewing sub-module 251, a data recommendation sub-module 252, a data search sub-module 254, an associated data sub-module 254, and the like.

The data adding/editing/viewing sub-module 251 is used to provide functions of adding, editing, viewing, etc. data.

The data recommendation sub-module 252 is used to recommend aggregate data to a user or data stored according to user image recommendations.

The data searching sub-module 253 is configured to obtain a search requirement of a user, and transmit the search requirement to the searching sub-module 243 in the data management module 240. The search sub-module 243 retrieves corresponding data from the database 230 according to the search requirement, and feeds back the retrieved data to the data search sub-module 253.

The association data sub-module 254 is used to associate or recommend the recorded data to other applications.

In some exemplary embodiments, as shown in fig. 4, a data set recommendation method is provided, including the following steps 402 to 406. Wherein:

step 402, a target tag is obtained.

The recording data may be structured information obtained by a data recording function of the electronic device and obtained after processing. The data recording function is used for storing the data information after analysis. Structured information is a form of data organization in which data is organized into predefined formats, which may include explicit fields or columns, and corresponding data types, etc., in such a way that the storage, retrieval, and analysis of the data becomes more efficient and accurate. In an actual application scene, the recorded data can be obtained by processing voice information input by a user through a data recording function, or can be obtained by processing text information input by the user through a data function, or can be obtained by processing interface information of an interface through the data function. After a piece of recorded data is acquired, the recorded data may be stored for review by the user as needed.

It is easy to understand that each piece of recorded data has a corresponding tag. The tag of the record data may be correspondingly generated when the record data is generated. Tags may be understood as semantic keywords of the corresponding record data. Each piece of recorded data may include one or more tags. The label of each record data can be generated by manual labeling or can be generated by semantic recognition of the record data through a semantic recognition algorithm. In the case where it is assumed that there is no duplicate recorded data, the tags of different recorded data are generally different. The target tags may be used to characterize content of interest to the user. After the target tag is determined, the recorded data of interest to the user may be acquired based on the target tag.

For example, the target tag may be determined according to the number of recorded data corresponding to the tags of all recorded data. For example, a tag whose number of recorded data exceeds the number threshold is determined as a target tag, or a tag whose number of recorded data is from high to low is determined as a target tag, or the like. Or the candidate data can be screened from the local data list according to preset conditions and stored in the candidate data list, and the target label is determined according to the number of record data matched with the candidate label of the candidate data in the local data list. Or the target tag may also be determined according to the user operation content.

And step 404, aggregating target data matched with the target labels into corresponding recommendation sets, wherein the target data are recorded from at least two application programs, and the target data are obtained based on a preset recording operation.

The aggregation is used for representing the classifying process according to the preset classifying rules. Each collection is used to characterize a class of recorded data. For example, the recorded data may be categorized according to transaction attributes and a corresponding collection of category data may be created, such as a corresponding collection of travel strategies, fitness, delicacies, or personal relationships. It is easy to understand that the number and types of the collections may be set according to the actual application scenario, and are not particularly limited herein.

For example, record data including a target tag in the local data list may be used as target data matching the target tag. In an actual application scene, if the target labels include a plurality of target data matched with each target label, selecting the target data with the number not greater than a first threshold value as aggregate data, wherein the total number of the aggregate data corresponding to each target label does not exceed a second threshold value, and the first threshold value is smaller than the second threshold value. The second threshold may be an integer multiple of the first threshold. For example, the first threshold is 100 and the second threshold is 1000. The first threshold or the second threshold may be set according to an actual application scenario, which is not limited herein. It is easy to understand that if there is a number of target data that matches the target tag that is greater than the first threshold, the first threshold number of target data may be selected as aggregate data from the target data that matches the target tag from near to far according to the data generation time. The aggregated data is then aggregated into corresponding recommended collections, whereby the number of aggregated data aggregated into a corresponding recommended collection at a time can be limited, reducing the storage pressure of the recommended collections.

For example, the target data may be matched with the album keywords of the recommended album, and the recommended album corresponding to the album keywords matched with the target data may be used as the recommended album corresponding to the target data. Wherein the collection keywords are used to characterize respective collections, each collection may include one or more collection keywords. It is easy to understand that the target data is matched with the recommended collection, and the semantic similarity between the label of the target data and the collection keyword of the recommended collection is higher than the similarity threshold.

Wherein the recorded data stored in the electronic device may be recorded from one or more applications installed on the electronic device. I.e. the data in each application on the electronic device can be recorded. The target data is recorded from at least two applications. For example, the target data includes data recorded from the a application and data recorded from the B application, or the target data includes data recorded from the a application, data recorded from the B application, data recorded from the C application, and the like. In some application scenarios, the at least two applications may include a target application having a data recording function.

The preset recording operation includes, for example, a screenshot operation, a praise operation, a sharing operation, a collection operation, or a search operation. The preset recording operation may also include an operation of triggering a voice conversation, or meeting other preset recording conditions, or the like. In an actual application scene, responding to the trigger of the preset recording operation, and recording the content corresponding to the preset recording operation to obtain corresponding recording data. Typically, the target data may be part or all of the recorded data held in the electronic device.

Step 406, displaying the recommendation set.

The aggregation of the target data into the corresponding recommendation set is equivalent to the process of updating the original recommendation set. After the updated set of recommendations is obtained, the set of recommendations may be displayed. The recommended collection may include a plurality of, for example, the plurality of recommended collections may be displayed in turn in a preset order, or the recommended collection may be displayed randomly. The recommended collection may be displayed in the form of a card, a pop-up window, a banner, a bullet screen, or the like, for example, and is not limited thereto.

For example, the recommendation sets may be displayed in a recommendation list of the corresponding application interface, e.g., each recommendation set may be dynamically displayed, or each recommendation set may be tiled, or each recommendation set may be displayed in a stack, etc.

According to the data collection recommendation method, the target label is obtained, the target data matched with the target label are aggregated into the corresponding recommendation collection, and the recommendation collection is displayed, so that the data matched with the real requirements of the user can be selected and aggregated into the collection and then recommended to the user, the recommendation accuracy is improved, meanwhile, the data information matched with the real requirements of the user is automatically provided for the user, and the use experience of the equipment can be improved.

In some embodiments, obtaining the target tag in step 402 includes:

and acquiring the target label according to the number of the record data matched with the candidate label of the candidate data in the candidate data list in the local data list.

Wherein the candidate data list is a list for storing candidate data. The candidate data are record data screened out from the local data list according to preset conditions. The candidate tag refers to a tag of candidate data. The record data in the local data list is used for representing all record data stored in the local of the electronic device, namely, the local data list stores the total data recorded and saved by the user. The preset condition may be, for example, a preset period, a preset theme, a preset location, or the like. It is easy to understand that different application scenarios, the corresponding preset conditions may be different. For example, in one application scenario, the preset condition may be a preset period, in another scenario, the preset condition may be a preset theme, or the like. Illustratively, the preset period includes a period of a last preset duration, such as a last day (24 hours), a last 2 days, or a last month, etc., which may be set according to an actual application scenario.

In an exemplary embodiment, the electronic device is pre-established with a candidate data list in which candidate data is stored and a local data list in which the full amount of data recorded by the user is stored. The electronic equipment can acquire candidate tags of candidate data, match each candidate tag with the record data in the local data list, acquire the number of record data matched by each candidate tag, and take the specified number of candidate tags with the number of matched record data from more to less as target tags. It is easy to understand that the greater the number of matching recorded data, the more frequently the corresponding candidate tag appears, the more capable of representing the user's true intent. Or the candidate labels with the matched record data quantity larger than the quantity threshold value can be used as target labels.

In this embodiment, the target tag is obtained according to the number of record data matched in the local data list by the candidate tag of the candidate data in the candidate data list, so that the target tag can be quickly determined from the candidate tags, and meanwhile, the accuracy of the target tag is ensured.

In some embodiments, obtaining the target tag according to the number of record data in the local data list that the candidate tag of the candidate data in the candidate data list matches, includes:

the method comprises the steps of obtaining candidate labels of candidate data in a candidate data list, matching record data in a local data list with the candidate labels, obtaining the number of the record data matched with each candidate label, and determining target labels with the number of the matched record data larger than a number threshold value from the candidate labels.

The labels of each record data in the local data list can be compared with the candidate labels in sequence, if the label of a certain record data comprises one of the candidate labels, the record data is matched with the corresponding candidate label, and the like, the record data matched with each candidate label and the number of matched record data can be obtained.

In an actual application scene, the electronic device may acquire a tag of each candidate data in the candidate data list, and perform deduplication processing on the tags of all the candidate data in the candidate data list to obtain candidate tags of the candidate data. And matching the record data in the local data list with each candidate tag to obtain record data matched with each candidate tag, acquiring the quantity of the record data matched with each candidate tag, comparing the quantity of the record data matched with each candidate tag with a quantity threshold value, and using the candidate tag with the quantity of the matched record data larger than the quantity threshold value as a target tag. Easily understood, the number of matched candidate labels with the recorded data being greater than the number threshold value indicates that the candidate labels are high in occurrence frequency, the intention of a user can be represented to a certain extent, and target data more meeting the requirements of the user can be screened based on the target labels. The number threshold may be set according to an actual application scenario. For example, the number threshold may be3, 5, 10, or the like.

In this embodiment, by matching the record data in the local data list with the candidate tags of the candidate data in the candidate data list, and determining the target tags with the number of matched record data greater than the number threshold from the candidate tags, the target tags matched with the real needs of the user can be rapidly determined, and the accuracy and the determination efficiency of the target tags can be improved.

In some embodiments, before obtaining the candidate tag of the candidate data in the candidate data list, the method further includes:

And executing the step of acquiring the candidate labels of the candidate data in the candidate data list in the case that the number of the candidate data reaches the target number.

Wherein the candidate data is record data selected from the local data list. If the number of candidate data is small, the number of target tags determined based on the candidate tags of the candidate data is small, and the number of target data matched based on the target tags is small, so that the execution times of the data set recommendation flow are increased, and the power consumption is increased. Meanwhile, the degree of updating the recommendation set is small every time, so that the recommendation set is easy to be high in repeatability, and the use experience of data set recommendation is reduced.

The electronic device may filter, in real time, data meeting a preset condition from the local data list, store the data as candidate data in the candidate data list, identify the number of candidate data in the candidate data list, and if the number of candidate data in the candidate data list reaches the target number, execute the step of acquiring the candidate tag of the candidate data in the candidate data list, that is, execute the data set recommendation procedure. The target number may be set according to an actual application scenario. For example, the target number is 5, 10, 20, or the like.

In an exemplary embodiment, the electronic device monitors the number of candidate data in the candidate data list, and may set the variable value of the number of candidate data in the candidate data list to 0 in an initial state, add 1 to the variable value of the number of candidate data to each time when one candidate data is added to the candidate data list until the variable value of the number becomes the target number, obtain candidate tags of the candidate data in the candidate data list, then match data in the local data list with the candidate tags, obtain the number of data matched with each candidate tag, determine, from the candidate tags, a target tag whose number of matched data is greater than the number threshold, or set, as a target tag, a specified number of candidate tags whose number of matched data is greater than the number of target tags, aggregate the target data matched with the target tags in the local data list to a corresponding recommendation set, and display the recommendation set.

In this embodiment, by identifying the number of candidate data in the candidate data list, and executing the step of acquiring the candidate tag of the candidate data in the candidate data list when the number of candidate data reaches the target number, the number of times of executing the data set recommendation procedure in a fixed time can be reduced, and the power consumption of the device can be reduced. Meanwhile, through setting of the target quantity, equipment power consumption and recommendation repeatability can be well balanced, and therefore equipment use experience is improved.

In some embodiments, after obtaining the candidate tag of the candidate data in the candidate data list, the method further comprises:

The candidate data of the acquired candidate tag is removed from the candidate data list.

In this embodiment, if the candidate tag of the candidate data has been acquired, it is indicated that the corresponding candidate data has been executed with the data set recommendation procedure, that is, the candidate tag of the candidate data has been involved in screening the target data, and the corresponding recommendation set is displayed, then the candidate data of the acquired candidate tag may be removed from the data list, so as to avoid repeated acquisition of the candidate tag of the candidate data, thereby affecting the accuracy of the recommendation set.

In some practical application scenarios, under the condition that the number of candidate data in the candidate data list is identified to reach the target number, the electronic device obtains candidate tags of the candidate data in the candidate data list, matches data in the local data list with the candidate tags, obtains the number of data matched with each candidate tag, determines target tags, of which the number of matched data is greater than a number threshold, from the candidate tags, and aggregates the target data matched with the target tags in the local data list into corresponding recommendation sets, so that the corresponding recommendation sets are displayed according to a given strategy. After the candidate tag of the candidate data in the candidate data list is acquired, the candidate data of the acquired candidate tag is removed from the candidate data list, i.e., is not saved in the candidate data list. The candidate data list is used for storing the candidate data which are newly screened out later, and when the number of the candidate data which are newly screened out reaches the target number, the candidate labels of the candidate data in the candidate data list under the current condition are obtained, so that the data set recommendation process can be repeatedly executed.

In this embodiment, after the candidate tag of the candidate data in the candidate data list is acquired, the candidate data of the acquired candidate tag is removed from the candidate data list, so that repeated acquisition and use of the candidate tag can be avoided, sufficient update of the candidate data in the candidate data list can be realized, and accuracy of a target tag determined from the candidate tag can be improved, so that recommendation accuracy of data set recommendation can be improved each time.

In some embodiments, aggregating the target data that matches the target tags into a corresponding recommendation set in step 404 includes:

Determining target data matched with the target tag from a local data list, performing de-duplication processing on the target data matched with the target tag to obtain de-duplicated target data, and aggregating the de-duplicated target data into a corresponding recommendation set.

Wherein the target tag may include one or more. If the tag of a certain target data comprises a target tag, the target data is matched with the included target tag. For example, assuming that the target tag includes tag 2, tag 4, and tag 5, the tag of the target data a includes tag 1, tag 2, and tag 3, since the tag of the target data a includes tag 2 in the target tag, it is explained that the target data a matches with tag 2 in the target tag. Assuming that the tag of the target data B includes the tag 2 and the tag 4, since the tag of the target data B includes the tag 2 and the tag 4 in the target tag, it is explained that the target data a matches with the tag 2 and the tag 4 in the target tag. Or if the similarity between the target data and the target label reaches the similarity threshold value, indicating that the target data is matched with the target label. Target data matched with each target tag can be respectively determined from the local data list, the target data matched with each target tag is subjected to de-duplication processing to obtain de-duplicated target data, and then the de-duplicated target data are aggregated into corresponding recommendation sets. The deduplication process refers to a process in which only one of the duplicated objects is retained. Since one piece of recorded data may include a plurality of tags, different target tags may match the same target data, so that repeated target data may occur, requiring deduplication processing of the target data.

The electronic device may determine, from the local data list, target data that matches each target tag, thereby obtaining target data that matches each target tag, and perform deduplication processing on the target data that matches each target tag, thereby obtaining deduplicated target data. For example, the repeated data can be screened from the target data matched with each target tag, if the repeated data exists, one piece of record data in the repeated data is reserved until the repeated record data does not exist in all the target data, and the target data after the repeated removal is obtained. And then aggregating the target data after the duplication removal into a corresponding recommendation set. The implementation of the deduplication process is not limited herein, and it is only necessary to implement that duplicate recorded data does not exist in the target data after deduplication.

In this embodiment, by determining the target data matched with the target tag from the local data list, performing deduplication processing on the target data matched with the target tag to obtain deduplicated target data, and then aggregating the deduplicated target data into a corresponding recommendation set, a situation that the finally matched target data is repeated due to the fact that different target tags match the same target data can be avoided, and the recommendation experience of the set of equipment is improved.

In some embodiments, aggregating the target data matching the target tags into a corresponding recommendation set in step 104 includes:

each item of target data matched with the target tag in the local data list is matched with the historical recommendation set, if the historical recommendation set is successfully matched with the target data, the target data is aggregated into the matched historical recommendation set, if the target data is not successfully matched with all the historical recommendation sets, a new recommendation set is created according to the target data, and the new recommendation set comprises the target data.

The historical recommendation collection refers to a recommendation collection which is already established. The historical recommendation set may be created based on the initially determined target data. The collection of historical recommendations may include one or more. It is easy to understand that if the data set recommendation process is performed for the first time, there may be no historical recommendation set, and then a corresponding recommendation set needs to be created according to the target data. Aggregating the target data into a matched set of historical recommendations may be understood as categorizing the target data into corresponding data categories.

The electronic device sequentially matches each item of target data matched with the target tag in the local data list with each historical recommendation set, and if the historical recommendation sets are successfully matched with the target data, aggregates the target data into the matched historical recommendation sets. And if the semantic similarity between the set keywords of the historical recommendation set and the target data is greater than a similarity threshold value, indicating that the target data is matched with the historical recommendation set. Or the collection keywords of the history recommendation collection can be matched with the labels of the target data, and if the semantic similarity between one collection keyword and one label of the target data is larger than a similarity threshold value, the corresponding history recommendation collection is matched with the target data. Otherwise, if the semantic similarity between the collection keywords of the historical recommendation collection and the target data is not greater than the similarity threshold value, or the semantic similarity between the collection keywords and the labels of the target data is not greater than the similarity threshold value, the historical recommendation collection is not matched with the target data.

For example, if the target data is not successfully matched with all the historical recommendation sets, it is indicated that the target data does not belong to a data category corresponding to any one of the historical recommendation sets, a new recommendation set needs to be created according to the target data, and the target data can be understood to be determined by initialization. In other words, the initialization-determined target data refers to target data used to create a new recommendation set. For example, if the target data is not successfully matched with all the historical recommendation sets, a set keyword may be determined according to the target data, and a new recommendation set corresponding to the target data may be created according to the set keyword. For example, a new recommended collection may be created using the determined collection keywords as collection names, and the target data may be added to the new recommended collection. As an example, if the determined album keyword is "all-place food", a new recommended album having an album name of "all-place food" may be created, and the target data may be added to the new recommended album having the name of "all-place food".

In some examples, if the target data and all the historical recommendation sets are not successfully matched, determining a set keyword according to the target data, screening first data matched with the determined set keyword from other target data which are not successfully matched with all the historical recommendation sets and subsequently appearing, under the condition that the number of the first data reaches a preset number, creating a new recommendation set corresponding to the set keyword based on the first data, and if the number of the first data does not reach the preset number, not creating the new recommendation set corresponding to the set keyword. Each recommendation set typically includes multiple entries of label data.

For example, in the case of creating a complete new collection of recommendations, the user may be supported to manually update the new collection of recommendations. For example, the user is supported to manually modify the collection names of the new recommendation collection, the user is supported to manually add other data information to the new recommendation collection, the user is supported to manually delete part of the data information from the new recommendation collection, and the like.

In this embodiment, each item of target data matched with the target tag in the local data list is matched with the historical recommendation set, if the historical recommendation set is successfully matched with the target data, the target data is aggregated into the matched historical recommendation set, if the target data is not successfully matched with all the historical recommendation sets, a new recommendation set is created according to the target data, and therefore all the target data matched with the target tag can be accurately aggregated into the corresponding recommendation set, and accuracy of the recommendation set is improved.

In some embodiments, if there is a successful match of the historical recommendation set with the target data, aggregating the target data into the matched historical recommendation set includes:

The method comprises the steps of calculating the similarity of each record data in a history recommendation set and the matched history recommendation set if the history recommendation set and the target data are successfully matched, aggregating the target data into the matched history recommendation set if the similarity of each record data and the target data in the matched history recommendation set is smaller than or equal to a similarity threshold value, and not aggregating the target data into the matched history recommendation set if the record data with the similarity of the target data being larger than the similarity threshold value exists in the matched history recommendation set.

The similarity of each piece of record data in the historical recommendation set and the target data is used for representing semantic similarity between each piece of record data in the historical recommendation set and the target data. If the similarity is smaller than or equal to the similarity threshold, the corresponding record data in the history recommendation set is indicated to be less similar to the target data, and dissimilar target data can be aggregated into the corresponding history recommendation set. The similarity threshold may be set according to the actual application scenario, for example, the similarity threshold is 80%, 85% or 90%, etc.

After determining a history recommendation set matched with the target data, the similarity between each piece of record data and the target data in the matched history recommendation set is calculated, the similarity between each piece of record data and the target data in the matched history recommendation set is compared with a similarity threshold value, if the similarity between each piece of record data and the target data in the matched history recommendation set is smaller than or equal to the similarity threshold value, the target data is aggregated into the matched history recommendation set, and if record data with the similarity greater than the similarity threshold value exists in the matched history recommendation set, the fact that the record data with the similarity similar to the target data exists in the matched history recommendation set is described, the target data can not be aggregated into the matched history recommendation set, namely the target data is regarded as repeated data, and repeated two pieces of data are prevented from being aggregated into the same recommendation set.

In this embodiment, by calculating the similarity between the target data and the record data in the matched historical recommendation set, under the condition that the similarity is smaller than or equal to the similarity threshold, that is, under the condition that the target data is not very similar to the record data in the matched historical recommendation set, the target data is aggregated into the matched historical recommendation set, and if the similarity between the record data and the target data exists in the historical recommendation set and is greater than the similarity threshold, that is, under the condition that the data similar to the target data exists in the matched historical recommendation set, the target data is not aggregated into the matched recommendation set, so that a plurality of data similar in semantic meaning can be prevented from being aggregated into the same recommendation set, recommendation resources are occupied, and recommendation efficiency is improved.

In some embodiments, obtaining the target tag in step 402 includes:

and acquiring user operation content, and acquiring the target label according to the user operation content.

The user operation content can be, for example, content that a user searches, browses, endorses, collects or downloads on a system application or a third party application, and the user operation content is used for representing content of interest to the user.

For example, the electronic device may detect the user operation content in real time and determine the target tag according to the detected user operation content within the target time period. The target duration may be set according to an actual application scenario. For example, the target time period is 1 hour, 2 hours, 5 hours, or the like. Alternatively, when the detected user operation content reaches the target amount, the target tag may be determined based on the detected user operation content.

For example, the electronic device may determine the target tag based on the semantics of the user-operated content.

In the embodiment, the target tag is acquired according to the user operation content, the target data matched with the target tag in the local data list is aggregated to the corresponding recommendation set, and the recommendation set is recommended to the user, so that the matching degree between the recommendation set and the real requirement of the user can be improved, and the accuracy of data set recommendation is improved.

In some embodiments, obtaining the target tag according to the user-operated content includes:

and determining content keywords according to the semantic recognition result, and determining the content keywords as target tags.

The electronic equipment can detect user operation content, extract semantic information of the user operation content, obtain a semantic identification result, characterize the semantic identification result through content keywords, and determine the content keywords as target tags. Content keywords may be used to characterize user-manipulated content.

In an actual application scene, the electronic device can detect the content of target operation performed by a user in a system application or a third party application in real time to obtain user operation content, semantic information of the user operation content is extracted through a semantic extraction tool to obtain a semantic recognition result, and the semantic recognition result is summarized into content keywords to obtain a target label. The target operation can be operations such as searching, browsing, praying, collecting or transferring sharing by a user. For example, the user searches for which "cuisines in the XX place" on the search engine, the electronic device may detect which "cuisines in the XX place" are the content of the search operation, perform semantic recognition according to the content of the corresponding search operation, obtain a semantic recognition result, determine that the content keyword is "XX place" or "cuisine" according to the semantic recognition result, and determine that the content keyword is "XX place" or "cuisine" as the target tag. As an example, if the target operation is a browsing operation, the user operation content may be determined according to the browsing duration, for example, the content with the browsing duration greater than or equal to the duration threshold may be used as the user operation content, and if the browsing duration is less than the duration threshold, the content may not be used as the user operation content. In other words, if the electronic device detects the browsing operation of the user, it needs to further detect the browsing duration corresponding to the browsing operation, and if the browsing duration is greater than or equal to the duration threshold, it determines the content of the browsing operation as the user operation content, otherwise, if the browsing duration is less than the duration threshold, it does not determine the content corresponding to the browsing operation as the user operation content even if the browsing operation is detected.

In an exemplary embodiment, the electronic device may integrate the user operation content detected in the target duration or the user operation content reaching the target amount, obtain the integrated operation content, perform semantic recognition on the integrated operation content, obtain a semantic recognition result, determine a content keyword according to the semantic recognition result, and determine the content keyword as the target tag. Wherein the target amount may be characterized by the amount of memory space occupied or by the complexity of the structure. The size of the target amount may be set according to the actual application scenario. As an example, the inductive integration may be implemented through an artificial intelligence model, for example, the artificial intelligence model generates user operation content to perform inductive integration to obtain integrated operation content, generates a summary corresponding to the integrated operation content, extracts content keywords based on the summary, and determines the extracted content keywords as target tags. The way of generalizing the integration is not limited to implementation by artificial intelligence models, but can be implemented by other ways, not limited herein. Or the semantic recognition can be directly carried out on the user operation content detected in the target duration or the user operation content reaching the target quantity, so as to obtain a semantic recognition result, and then the content keywords are determined according to the semantic recognition result.

In the embodiment, the semantic recognition is performed on the user operation content to obtain the semantic recognition result, the content keyword is determined according to the semantic recognition result, and the content keyword is determined to be the target label, so that the target label is matched with the semantic of the user operation content, the target label is accurately determined through the user operation content, the matching between the target label and the user intention is improved, and the recommendation accuracy is improved.

In some embodiments, displaying the collection of recommendations in step 406 includes:

and displaying the recommendation sets in turn according to a preset sequence.

Typically, the recommended collection includes a plurality of sets, each recommended collection including a plurality of pieces of record data. The record data in each recommended collection can be arranged from new to old or from old to new according to the generated time sequence, or can be arranged according to other sequences, and the preset arrangement sequence of the plurality of record data in each collection can be set according to the actual application scene. Each displayed recommended collection can correspondingly display information such as collection names of the recommended collection, the number of included record data and the like. The preset sequence can be set according to the actual application scene, and the preset sequence comprises a random sequence, namely, a recommendation set is randomly displayed. Or the preset order may be an update time order of the recommended collection, for example, the preset order is an order in which the recommended collection update time is from late to early, the recommended collection updated later is arranged before, and the recommended collection updated earlier is arranged after.

For example, the recommended collections may be displayed in turn in a preset order, with a preset number of recommended collections displayed each time, and a preset duration displayed continuously. The preset number or preset duration may be set according to the actual application scenario, for example, the preset number is 3,4, or 5, and the preset duration is 1 second, 2 seconds, or 3 seconds, which is not limited herein. As an example, 3 recommendation sets are displayed each time in a preset order, 3 recommendation sets displayed each time last for 1 second, and then the other 3 recommendation sets are replaced for display, so that dynamic display of the recommendation sets is realized.

In an exemplary embodiment, in a process of alternately displaying the recommendation sets according to a preset sequence, in response to a user selection operation on the target recommendation set, the electronic device may display a certain amount of data at a time according to an arrangement manner of data in the target recommendation set, and display summary information of the target recommendation set. The number of the record data displayed each time can be set according to the actual application scene. The target recommendation set may be any recommendation set. Each piece of data in the target recommendation set can display at least one data information of corresponding recorded data name, recorded data source, associated data quantity, recorded data abstract, recorded data recording time, representative picture and the like. The arrangement of the data information is not particularly limited herein.

As an example, a schematic display of the recommendation set is shown in fig. 5. The recommended collection sets can be displayed on the application interface of the target application in turn according to a preset sequence, and each recommended collection set displays collection set names and the recorded data quantity included in the recommended collection set. In response to the selection operation of the target recommendation set "XX travel attack", as shown in fig. 6, the record data information included in the target recommendation set may be sequentially displayed according to a preset arrangement sequence in a list, stack or tile form, and information such as a record data name, a record data source, an associated data number, a record data abstract and a representative picture of each record data may be displayed. The name and summary information of the target recommendation set may also be displayed.

In this embodiment, by displaying the recommendation sets in turn according to a preset sequence, information of each recommendation set can be better displayed to a user, and data set recommendation experience of the device is improved.

In some embodiments, displaying the collection of recommendations in step 406 includes:

The stack displays the collection of recommendations.

The stacking display refers to a mode of displaying a plurality of recommendation sets in a staggered stacking mode. In an actual application scenario, the recommendation sets may be displayed in a stacked manner in the order from new to old or from old to new. Or the respective recommended collections may be displayed in a stack from high to low in the level of their highlighting. Wherein the level of sophistication of each recommended collection may be determined based on collection keywords of the recommended collection. For example, if the collection keywords represent topics such as sports, parties, travel, etc., the representation of the topic is high, and if the collection keywords represent topics such as schedule to be done, work memo, etc., the representation of the topic is low. It is easy to understand that the degree of the highlighting may be determined according to the actual application scenario.

In the embodiment, the recommendation collection is displayed in a stacked manner, so that the whole information of the recommendation collection can be visually displayed, and the recommendation experience of the data collection is improved.

In some embodiments, the above method further comprises:

And adding or deleting the recommended collection in response to the triggering operation of the collection editing control.

The collection editing control is a control for editing the recommended collection. Each recommended collection corresponds to a corresponding collection editing control, and the corresponding recommended collection can be added or deleted in response to the triggering operation of the collection editing control. The collection editing control at least comprises an adding control and a deleting control.

Illustratively, the target recommendation set is added to the My set in response to a trigger to an add control to the target recommendation set. And deleting the target recommendation set from the recommendation set list in response to triggering of a deletion control of the target recommendation set.

In some examples, the collection editing control further includes a change control. For example, the name of the displayed target recommendation set may be changed on the display interface of the recommendation set, or record data not interested in the target recommendation set may be deleted, or record data may be newly added to the target recommendation set, or the displayed target recommendation set may be moved to other set categories, for example, to "my set" which characterizes the set category focused by the user, or the target recommendation set may be labeled with a preset label, for example, "interested", "not interested", "no recommendation", etc., or the target recommendation set may be deleted. The target recommendation set may be any of the displayed recommendation sets.

An edit control corresponding to the edit operation of the recommendation collection is set on a display interface of the recommendation collection, and processing corresponding to the edit operation of the displayed recommendation collection is performed in response to triggering of the edit control. As an example, in response to a selection operation on the target recommendation set, record data information included in the target recommendation set is displayed in a preset arrangement order, and editing controls such as "reserved set", "uninteresting", and "chat" may be displayed. Wherein the "keep collection" control is used to move the target recommended collection into the "my collection" category; the "not interested" control is used for representing that the corresponding target recommendation collection is not interested in the current situation, and the "not recommending the collection any more", and the "chat" control is used for representing that the response or the processing is carried out according to the input content or the problem. It is easy to understand that the editing control can be set according to the actual application scenario. In the actual application scenario, if the "keep collection" control in fig. 6 is triggered, as shown in fig. 7, "added to my collection" is displayed, which indicates that the target recommended collection is added to the "my collection". The electronic device may also obtain, from the dialog box, a question related to the target recommendation set entered by the user, and display an answer corresponding to the question in the dialog box in response to a trigger of the chat control by the user. If, in response to the triggering of the "no interest" control, as shown in FIG. 8, the "no longer recommends this collection" is displayed.

In this embodiment, by displaying a plurality of recommended collections and a collection editing control, and adding or deleting a recommended collection in response to a triggering operation of the collection editing control, deletion or addition processing can be flexibly and conveniently performed on the recommended collection, and editing efficiency of the displayed recommended collection is improved, so that collection recommendation experience of the device is improved.

In some embodiments, the above method further comprises:

The method comprises the steps of responding to a viewing operation of a target recommendation collection, displaying a collection page of the target recommendation collection, responding to a triggering operation of a target control of the target page of the target recommendation collection, displaying candidate function controls, responding to the triggering operation of the target function controls, and executing functions corresponding to the target function controls.

Wherein the target recommendation set is any recommendation set. The target control is a control for triggering the display of candidate functionality controls. The target functionality control is one of the candidate functionality controls. Each candidate functionality control corresponds to a collection of processing functions. For example, functions of adding record data, batch management of record data in a target recommended collection, modifying collection names, adjusting collection rules, or deleting collections, and the like.

In one example, as shown in FIG. 9, the view operation includes a click operation. In response to a click operation on the target recommendation set, a set page 902 of the target recommendation set is displayed, in response to a click operation on a target control 904 of the target page of the target recommendation set, a candidate function control 906 is displayed, and in response to a click operation on a target function control 908, a function corresponding to the target function control is executed.

In this embodiment, by responding to the viewing operation of the target recommendation collection, displaying the collection page of the target recommendation collection, responding to the triggering operation of the target control of the target page of the target recommendation collection, displaying the candidate function control, responding to the triggering operation of the target function control, executing the function corresponding to the target function control, and being capable of conveniently realizing the processing of the recorded data in the target recommendation collection and improving the recommendation experience of the data collection.

In some embodiments, the above method further comprises:

The method comprises the steps of displaying a display interface of a target application program, displaying a question and answer assistant entrance on the display interface, recording data by the target application program based on a preset recording operation, displaying a question and answer assistant interface in response to a triggering operation on the question and answer assistant entrance, and displaying a question and answer result corresponding to the question and answer information in response to the question and answer information input in the question and answer assistant interface.

The target application program is an application program for recording data and displaying the recommended collection. The target application may record data based on a preset recording operation. The preset recording operation includes, for example, a screenshot operation, a praise operation, a sharing operation, a collection operation, or a search operation. The preset recording operation may also include an operation of triggering a voice conversation, or meeting other preset recording conditions, or the like.

For example, the electronic device may display a question and answer assistant interface in response to a long press operation, a click operation, or a trigger operation of a preset voice input or the like to the question and answer assistant entry, and may search for a question and answer result corresponding to the question and answer information from the local record list in response to the question and answer information input in the question and answer assistant interface, and display the question and answer result. Or may search for a question and answer result corresponding to the question and answer information from the internet in response to the question and answer information input in the question and answer assistant interface, and display the question and answer result. The question and answer assistant interface may call an artificial intelligence question and answer assistant, and the question and answer result corresponding to the question and answer information is determined through the artificial intelligence question and answer assistant. It is readily understood that the artificial intelligence question-answering assistant may be implemented by an artificial intelligence question-answering model or a question-answering algorithm, etc.

In the embodiment, the question and answer assistant interface is displayed in response to the triggering operation of the question and answer assistant entrance displayed on the display interface of the target application degree, the question and answer result corresponding to the question and answer information is displayed in response to the question and answer information input in the question and answer assistant interface, the question and answer can be conveniently realized, the question and answer result corresponding to the question and answer information is obtained, and the equipment use experience is improved.

In one example, a flow diagram of a data set recommendation method is shown in FIG. 10. Including the following steps 1002 through 1012.

Step 1002, triggering a data set recommendation process, and obtaining all A2 tags of the record data in the queue.

Wherein, the data set recommendation flow can be implemented when the target application program runs. The target application is an application to which the data set recommendation method in each of the above embodiments is loaded. The queue corresponds to a candidate data list, the data in the queue corresponds to candidate data, and all the A2 tags of the data in the queue correspond to all the candidate tags of the candidate data. I.e. all candidate labels of the candidate data in the candidate data list are obtained. The candidate data list is, for example, data generated during a period of a latest preset duration. Candidate data may be screened from a local data list. In an actual application scenario, the recommendation may be triggered when the number of candidate data in the candidate data list reaches 5 (target number), and the recommendation is not triggered when the number of candidate data in the candidate data list is less than 5. It is easily understood that when generating the recorded data, the corresponding tag can be generated accordingly.

Step 1004, retrieving local data, and counting the data quantity matched with each A2 label.

Wherein the local data corresponds to the record data in the local data list. Namely, matching the data in the local data list with the candidate tags to obtain the number of record data matched with each candidate tag.

In step 1006, A2 tags with a number of matching record data greater than 3 are screened out.

In this example, the number threshold is 3, that is, the target tag whose number of matched record data is greater than 3 is selected from the candidate tags.

Step 1008, filtering the record data in the local data list according to the screened A2 tags.

Wherein the selected A2 tag is equivalent to the target tag. I.e. the target data matched with the target tag is screened out from the local data list. In the screening process, the number of the target data matched with each target label does not exceed a first threshold value, and the number of the target data matched with all the target labels does not exceed a second threshold value. The present example illustrates that the first threshold value is 100 and the second threshold value is 1000. If the number of the target data matched with a certain target tag exceeds a first threshold value, selecting the target data of the first threshold value from the matched target data according to the data generation time from near to far, and taking the target data as the finally screened target data.

In step 1010, the request server generalizes the collection.

The electronic device may send a data aggregation request to the cloud or the server, so as to aggregate (generalize) the screened target data into corresponding recommendation sets.

Step 1012, store the returned results and place them in the recommendation set queue.

The returned result may be aggregation information of the target data returned by the server or the cloud, that is, association information of each item of target data and the corresponding recommendation set. Or the returned result may be a recommendation set returned by the server or the cloud to aggregate the target data. If the historical recommendation collection set matched with the target data does not exist, a new recommendation collection set is created according to the target data, and a collection set name of the new recommendation collection set is generated according to a collection set creation algorithm. I.e. the recommendation set comprises a history recommendation set and a new recommendation set.

After receiving the returned result, the electronic device may put the returned result into the recommendation set queue, and then display each recommendation set in turn according to a preset sequence. For example, the recommended collection is carousel-played N times a day, and at most a preset number of recommended collections are displayed each day, for example, at most 10 recommended collections are displayed each day. N is a positive integer, and can be set according to an actual application scene. The user can perform editing operations such as deleting, updating and the like on the displayed recommendation set.

In the above embodiment, by matching the record data in the local data list with the candidate tags of the candidate data in the queue, and using the candidate tag with the number of record data matched with the candidate tag being greater than the number threshold as the target tag, the target tag matched with the real requirement corresponding to the user history record data can be quickly determined, so that the corresponding data is filtered based on the target tag to aggregate, the aggregated recommendation set is recommended, automatic recommendation of the user on the data set is realized, and the recommendation accuracy can be improved, thereby greatly improving the recommendation experience of the data set of the device.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides a data set recommendation device for realizing the data set recommendation method. The implementation of the solution provided by the apparatus is similar to the implementation described in the above method, so the specific limitation in the embodiments of the data set recommendation apparatus provided below may be referred to the limitation of the data set recommendation method hereinabove, and will not be repeated herein.

In some exemplary embodiments, as shown in FIG. 11, a data collection recommendation apparatus 1100 is provided, comprising a tag acquisition module 1102, a data aggregation module 1104, and a collection display module 1106, wherein:

A tag obtaining module 1102, configured to obtain a target tag;

A data aggregation module 1104, configured to aggregate target data matched with the target tag into a corresponding recommendation set, where the target data is recorded from at least two applications;

The collection display module 1106 is configured to display the recommended collection.

In some embodiments, the tag obtaining module 1102 is further configured to obtain the target tag according to the number of record data matched in the local data list by the candidate tag of the candidate data in the candidate data list.

In some embodiments, the tag obtaining module 1102 is further configured to obtain candidate tags of candidate data in the candidate data list, match record data in the local data list with the candidate tags, obtain a number of record data matched with each candidate tag, and determine a target tag from the candidate tags, where the number of matched record data is greater than a number threshold.

In some embodiments, the apparatus further comprises a data quantity monitoring module for identifying a quantity of candidate data in the candidate data list before obtaining the candidate labels of the candidate data in the candidate data list, and executing the obtaining of the candidate labels of the candidate data in the candidate data list if the quantity of candidate data reaches the target quantity.

In some embodiments, the apparatus further includes a data removal module configured to remove candidate data of the acquired candidate tag from the candidate data list after acquiring the candidate tag of the candidate data in the candidate data list.

In some embodiments, the data aggregation module 1104 is further configured to determine target data matching the target tag from the local data list, perform deduplication processing on the target data matching the target tag to obtain deduplicated target data, and aggregate the deduplicated target data into a corresponding recommendation set.

In some embodiments, the data aggregation module 1104 is further configured to match each item of target data in the local data list that matches the target tag with the historical recommendation set, aggregate the target data into the matched historical recommendation set if there is a successful match between the historical recommendation set and the target data, and create a new recommendation set according to the target data if the target data is not successfully matched with all the historical recommendation sets, where the new recommendation set includes the target data.

In some embodiments, the data aggregation module 1104 is further configured to calculate a similarity of the target data to each of the record data in the matched set of history recommendations if there is a successful match between the set of history recommendations and the target data;

If record data with the similarity larger than the similarity threshold value exists in the matched historical recommendation collection, the target data is not aggregated into the matched historical recommendation collection.

In some embodiments, the tag obtaining module 1102 is further configured to obtain user operation content, and obtain a target tag according to the user operation content.

In some embodiments, the tag obtaining module 1102 is further configured to perform semantic recognition on the user operation content to obtain a semantic recognition result, determine a content keyword according to the semantic recognition result, and determine the content keyword as the target tag.

In some embodiments, the collection display module 1106 is further configured to alternately display the recommended collection in a preset order.

In some embodiments, the apparatus further comprises a collection editing module for displaying a plurality of recommended collections and a collection editing control, and adding or deleting recommended collections in response to a triggering operation of the collection editing control.

In some embodiments, the device further comprises a collection processing module, a trigger operation response to the target control of the target page of the target recommendation collection, a candidate function control display, and a function corresponding to the target function control execution response to the trigger operation to the target function control.

In some embodiments, the device further comprises a question and answer display module, wherein the question and answer display module is used for displaying a display interface of a target application program, a question and answer assistant entry is displayed on the display interface, the target application program is used for recording data based on a preset recording operation, the question and answer assistant interface is displayed in response to a triggering operation on the question and answer assistant entry, and a question and answer result corresponding to the question and answer information is displayed in response to the question and answer information input in the question and answer assistant interface.

The respective modules in the data set recommendation device may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or independent of a processor in the electronic device, or may be stored in software in a memory in the electronic device, so that the processor may call and execute operations corresponding to the above modules.

In some exemplary embodiments, an electronic device is provided, comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the data set recommendation method in the above embodiments when the computer program is executed.

In some embodiments, a computer readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the steps of the data collection recommendation method in the above embodiments.

In some embodiments, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the data collection recommendation method of the above embodiments.

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are both information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to meet the related regulations.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile memory and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (RESISTIVE RANDOM ACCESS MEMORY, reRAM), magneto-resistive Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computation, an artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) processor, or the like, but is not limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the present application.

The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims (19)

1. A data set recommendation method, the method comprising:

obtaining a target label;

aggregating target data matched with the target tag into a corresponding recommended collection, wherein the target data is recorded from at least two application programs, and the target data is obtained based on a preset recording operation;

And displaying the recommendation set.

2. The method of claim 1, wherein the obtaining the target tag comprises:

and acquiring the target label according to the number of the record data matched with the candidate label of the candidate data in the candidate data list in the local data list.

3. The method according to claim 2, wherein the obtaining the target tag according to the number of record data matched by the candidate tag of the candidate data in the candidate data list in the local data list includes:

Acquiring candidate labels of candidate data in a candidate data list;

Matching the record data in the local data list with the candidate tags, and obtaining the number of record data matched with each candidate tag;

And determining target tags, of which the matched recorded data quantity is greater than a quantity threshold, from the candidate tags.

4. A method according to claim 3, wherein prior to the obtaining of candidate tags for candidate data in the candidate data list, the method further comprises:

identifying a number of candidate data in a candidate data list;

and executing the step of acquiring the candidate labels of the candidate data in the candidate data list under the condition that the number of the candidate data reaches the target number.

5. A method according to claim 3, wherein after the candidate tags for candidate data in the candidate data list are obtained, the method further comprises:

and removing the candidate data of the acquired candidate tag from the candidate data list.

6. The method of claim 1, wherein the aggregating the target data matching the target tags into respective recommendation sets comprises:

Determining target data matched with the target tag from a local data list;

Performing de-duplication treatment on target data matched with the target tag to obtain de-duplicated target data;

And aggregating the target data subjected to the de-duplication into a corresponding recommendation set.

7. The method of claim 1, wherein the aggregating the target data matching the target tags into respective recommendation sets comprises:

Matching each item of target data matched with the target tag in the local data list with a historical recommendation set;

If the history recommendation collection is successfully matched with the target data, the target data is aggregated into the matched history recommendation collection;

If the target data are not successfully matched with all the historical recommendation collection sets, a new recommendation collection set is created according to the target data, and the new recommendation collection set comprises the target data.

8. The method of claim 7, wherein if there is a successful match of the historical recommendation set with the target data, aggregating the target data into a matched historical recommendation set, comprising:

if the history recommendation set is successfully matched with the target data, calculating the similarity of the target data and each piece of record data in the matched history recommendation set;

If the similarity between each piece of recorded data in the matched historical recommendation set and the target data is smaller than or equal to a similarity threshold value, aggregating the target data into the matched historical recommendation set;

and if the record data with the similarity larger than the similarity threshold value with the target data exists in the matched historical recommendation collection, not aggregating the target data into the matched historical recommendation collection.

9. The method of claim 1, wherein the obtaining the target tag comprises:

and acquiring user operation content, and acquiring a target label according to the user operation content.

10. The method of claim 9, wherein the obtaining the target tag according to the user operation content comprises:

carrying out semantic recognition on the user operation content to obtain a semantic recognition result;

And determining content keywords according to the semantic recognition result, and determining the content keywords as target tags.

11. The method of any of claims 1 to 10, wherein the displaying the collection of recommendations comprises:

And displaying the recommendation collection in turn according to a preset sequence.

12. The method of any of claims 1 to 10, wherein the displaying the collection of recommendations comprises:

The stack displays the collection of recommendations.

13. The method according to claim 1, wherein the method further comprises:

Displaying a plurality of recommended collection and collection editing controls;

and responding to the triggering operation of the collection editing control, and adding or deleting the recommended collection.

14. The method according to claim 1, wherein the method further comprises:

responding to a viewing operation of a target recommendation collection, and displaying a collection page of the target recommendation collection;

responding to the triggering operation of the target control of the target page of the target recommendation collection, and displaying candidate function controls;

and responding to the triggering operation for the target function control, and executing the function corresponding to the target function control.

15. The method according to claim 1, wherein the method further comprises:

the system comprises a display interface for displaying a target application program, wherein a question and answer assistant entrance is displayed on the display interface;

responding to the triggering operation of the question and answer assistant entrance, and displaying a question and answer assistant interface;

In response to the question and answer information input in the question and answer assistant interface, a question and answer result corresponding to the question and answer information is displayed.

16. A data set recommendation device, the device comprising:

the label acquisition module is used for acquiring the target label;

the data aggregation module is used for aggregating target data matched with the target tag into corresponding recommendation aggregate, wherein the target data is recorded from at least two application programs, and the target data is obtained based on a preset recording operation;

and the collection display module is used for displaying the recommended collection.

17. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 15 when the computer program is executed.

18. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 15.

19. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method of any one of claims 1 to 15.