CN110162796B - Method and device for creating news topics - Google Patents

Abstract

The purpose of the embodiment of the present application is to provide a method and device for creating a news topic, wherein the method includes: obtaining multiple pieces of news information provided by an information provider, clustering the multiple pieces of news information through a pre-trained information clustering model, Get multiple information collections, each information collection contains multiple news information corresponding to the same news event, through the news topic creation system, according to the multiple news information corresponding to each news event, create corresponding news topics for each news event, news Topics are used to report corresponding news events. Through the embodiments of the present application, the efficiency of creating news topics can be improved.

Description

Method and device for creating news topics

technical field

The present application relates to the field of computers, in particular to a method and device for creating news topic.

Background technique

At present, the main process of pushing news topics to users is that after obtaining multiple news information from various news providers, the staff browses the content of these news information, and manually screens the news information combinations that meet the preset content requirements according to the content of these news information Create news topics and push news topics to users. Wherein, each news topic is used to report a news event, and each news topic includes a topic name and a plurality of news information.

It can be seen that the current method of manually combining news information into news topics has the problem of low efficiency in creating news topics.

Contents of the invention

The purpose of the embodiments of the present application is to provide a method and device for creating news topics, so as to improve the efficiency of creating news topics.

In order to solve the above technical problems, the embodiment of the present application is implemented as follows:

The embodiment of this application provides a method for creating a news topic, including:

Obtain multiple news information provided by the information provider, and the news information corresponding to the news event is recorded in the news information;

The multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets; wherein, the news content of the multiple pieces of news information included in each information set corresponds to the same news event, and different The news events corresponding to the above information sets are different;

Through the news topic creation system, according to the multiple news information corresponding to each of the news events, a corresponding news topic is created for each of the news events; wherein, the news topic is used to report the corresponding news event; the news A topic includes a topic name and topic information, and the topic information is determined based on a plurality of pieces of news information corresponding to news events reported in the news topic.

The embodiment of this application provides a method for creating a news topic, including:

Obtain multiple news information provided by the information provider, and the news information corresponding to the news event is recorded in the news information;

The multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets; wherein, the news content of the multiple pieces of news information included in each information set corresponds to the same news event, and different The news events corresponding to the above information sets are different;

According to a plurality of pieces of news information corresponding to each of the news events, determine the event heat of each of the news events, and select the target event in each of the news events according to the event heat of each of the news events;

Create a news topic for the target event through the news topic creation system according to multiple news information corresponding to the target event; wherein, the news topic is used to report the target event; the news topic includes topic name and topic Information, the topical information is determined according to multiple pieces of news information corresponding to the target event.

An embodiment of the present application provides a device for creating news topics, including:

The first obtaining module is used to obtain a plurality of news information provided by the information provider, and the news content corresponding to the news event is recorded in the news information;

The first clustering module is used to cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information sets; wherein, each of the information sets includes the news content of multiple pieces of news information Corresponding to the same news event, the news events corresponding to different information sets are different;

The first creation module is used to create corresponding news topics for each of the news events according to the multiple pieces of news information corresponding to each of the news events through the news topic creation system; wherein, the news topics are used to report the corresponding news topics The news event is mentioned; the news topic includes a topic name and topic information, and the topic information is determined according to a plurality of pieces of news information corresponding to the news event reported in the news topic.

An embodiment of the present application provides a device for creating news topics, including:

The second obtaining module is used to obtain a plurality of news information provided by the information provider, and the news content corresponding to the news event is recorded in the news information;

The second clustering module is used to cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information sets; wherein, each of the information sets includes news content of multiple pieces of news information Corresponding to the same news event, the news events corresponding to different information sets are different;

The selection module is used to determine the event heat of each of the news events according to a plurality of news information corresponding to each of the news events, and select the target event in each of the news events according to the event heat of each of the news events;

The second creation module is used to create a news topic for the target event according to multiple pieces of news information corresponding to the target event through the news topic creation system; wherein, the news topic is used to report the target event; the A news topic includes a topic name and topic information, and the topic information is determined based on a plurality of pieces of news information corresponding to the target event.

An embodiment of the present application provides a device for creating a news topic, including: a processor; and a memory arranged to store computer-executable instructions, and when the computer-executable instructions are executed, the processor implements the above-mentioned news topic Steps to create a method.

The embodiment of the present application provides a storage medium for storing computer-executable instructions, and when the computer-executable instructions are executed, the steps of the above-mentioned method for creating a news topic are realized.

In the embodiment of the present application, first obtain multiple pieces of news information provided by the information provider, and then cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information sets, each information set includes multiple The news content of each piece of news information corresponds to the same news event, and finally, through the news topic creation system, a corresponding news topic is created for each news event according to the multiple news pieces corresponding to each news event. It can be seen that through the embodiment of the present application, news topics can be automatically created through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency of manually creating news topics and improving the efficiency of creating news topics.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments described in this application. For those skilled in the art, other drawings can also be obtained according to these drawings on the premise of not paying creative efforts.

FIG. 1 is a schematic diagram of a scene of a method for creating news topics provided by an embodiment of the present application;

FIG. 2 is a schematic flowchart of a method for creating a news topic provided by an embodiment of the present application;

Fig. 3a is a schematic interface diagram of a news topic creation system provided by an embodiment of the present application;

Fig. 3b is a schematic interface diagram of a news topic creation system provided by another embodiment of the present application;

Fig. 3c is a schematic interface diagram of a news feature creation system provided by another embodiment of the present application;

FIG. 4 is a schematic diagram of an interface of a user terminal provided by an embodiment of the present application;

FIG. 5 is a schematic flowchart of a method for creating news topics provided by another embodiment of the present application;

FIG. 6 is a schematic diagram of the module composition of a news topic creation device provided by an embodiment of the present application;

FIG. 7 is a schematic diagram of module composition of a news feature creation device provided by another embodiment of the present application;

Fig. 8 is a schematic structural diagram of a device for creating a news topic according to an embodiment of the present application.

Detailed ways

In order to enable those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described The embodiments are only some of the embodiments of the present application, but not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

The purpose of the embodiments of the present application is to provide a method and device for creating news topics, so as to improve the efficiency of creating news topics. The method for creating a news topic in the embodiment of the present application may be executed by a news topic creation device (such as a background server).

Figure 1 is a schematic diagram of a scene of a method for creating a news feature provided by an embodiment of the present application. As shown in Figure 1, the scene includes a user terminal and a news feature creation device, wherein the user terminal includes but is not limited to a tablet as shown in Figure 1 Computer 101 , mobile phone 102 , desktop computer 103 , notebook computer 104 , and news feature creation equipment include but not limited to server 200 as shown in FIG. 1 . In this scenario, the news topic creation device can execute the news topic creation method provided in the embodiment of the application to generate a news topic, and push the news topic to the user terminal, and the user terminal can display the news topic.

Fig. 2 is the schematic flow chart of the news topic creation method that an embodiment of the present application provides, as shown in Fig. 2, this flow process comprises the following steps:

Step S202, obtaining a plurality of news information provided by the information provider, and the news content corresponding to the news event is recorded in the news information;

Step S204, clustering the multiple pieces of news information through the pre-trained information clustering model to obtain multiple information collections; wherein, the news content of the multiple news information included in each information collection corresponds to the same news event, different information The news events corresponding to the sets are different;

Step S206, through the news feature creation system, according to the news information corresponding to each news event, create a corresponding news feature for each news event; wherein, the news feature is used to report the corresponding news event; the news feature includes the feature name and feature information , the special topic information is determined according to multiple pieces of news information corresponding to the news events reported by the news special topic.

In the embodiment of the present application, first obtain multiple pieces of news information provided by the information provider, and then cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information sets, each information set includes multiple The news content of each piece of news information corresponds to the same news event, and finally, through the news topic creation system, a corresponding news topic is created for each news event according to the multiple news pieces corresponding to each news event. It can be seen that through the embodiment of the present application, news topics can be automatically created through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency of manually creating news topics and improving the efficiency of creating news topics.

In the above step S202, a plurality of news information provided by the information provider is obtained, and the news content corresponding to the news event is recorded in the news information. specifically. The information provider can be various news providers such as China Business News and Central News, and the news topic creation equipment obtains the news information provided by each information provider. The news content corresponding to the news event is recorded in the news information. For example, the news event is the 11th military parade event. News information A can record the news content about the inspected soldiers in the 11th military parade event. News information B can record There are news content about the inspected weapons and equipment in the eleventh military parade.

In the above-mentioned step S204, the multiple pieces of news information acquired are clustered through the pre-trained information clustering model to obtain multiple information collections. Wherein, the news content of multiple news information included in each information collection corresponds to the same news event, and the news events corresponding to different information collections are different.

For example, through step S202, ten pieces of news information are obtained, among which three pieces of news information respectively record the relevant news content of the 11th military parade, the other three pieces of news information respectively record the relevant news content of the Belt and Road Summit, and the other four pieces of news information respectively Records related news content of the Dragon Boat Festival celebrations. Then in step S204, these ten pieces of news information are clustered through the pre-trained information clustering model, and three pieces of news information can be obtained. Among them, the three pieces of news information included in the first information set all record information about the 11th military parade. For related news content, the three news information included in the second information collection all record the relevant news content of the Belt and Road Summit, and the four news information included in the third information collection all record the relevant news content of the Dragon Boat Festival celebration.

The specific process of the above step S204 may be:

(a1) extract the information titles of each news information by the pre-trained information clustering model, and calculate the semantic similarity between the information titles of each news information;

(a2) Through the pre-trained information clustering model, according to the semantic similarity between the information titles of each news information, each news information is clustered to obtain multiple information sets.

Specifically, the pre-trained information clustering model can be realized through clustering algorithms such as Affinity Propagation. In this embodiment, the information clustering model extracts the information titles of each piece of news information, and performs word segmentation processing on each information title, and determines the word vector of the word obtained by each word segmentation. For an information title, according to the information in the information title The word vector of each word determines the vector of the information title, calculates the distance between each information title according to the vector of each information title, and determines the semantic similarity between each information title according to the distance between each information title. Of course, the information clustering model can also use other existing algorithms to determine the semantic similarity between the information titles of each news information, which is not limited here.

After determining the semantic similarity between the information titles of each piece of news information, the information clustering model can group the information titles with high semantic similarity into one category, and the clustering center can be flexible according to the increase of information titles in this category. Adjustment. After clustering each information title into multiple categories, the information clustering model clusters the news information corresponding to each information title according to the clustering method, so as to cluster each news information and obtain multiple information sets. In this embodiment, the information clustering model can also be realized by other existing algorithms, which is not limited here.

One training method of the information clustering model is: manually set a large number of sample information in advance, and mark the information title of each sample information, and manually cluster each information title into multiple categories, and cluster each sample information into multiple categories, Input the sample information, the information title of each sample information, the clustering method of each information title, and the clustering method of each sample information into the neural network model for training. The neural network model trained by these sample data is the information clustering Model.

It can be seen that through this embodiment, the pre-trained information clustering model can be used to cluster multiple pieces of news information obtained to obtain multiple information sets. It can be understood that each information set corresponds to a news event, so as to achieve event mining. Effect.

In this embodiment, after clustering the multiple pieces of news information obtained through the pre-trained information clustering model to obtain multiple information sets, an event name can also be generated for the news event corresponding to each information set. The process is specifically as follows: for each information collection, word segmentation is performed on the information titles of the news articles in the information collection, and the event name of the news event corresponding to the information collection is determined according to the multiple words obtained by the word segmentation.

For example, if two information sets are obtained by co-clustering, then for one of the information sets, word segmentation processing is performed on the information titles of the news information in the information set, and the news corresponding to the information set is determined according to the multiple words obtained by word segmentation. The event name of the event, and then, for another information collection, perform word segmentation processing on the information titles of each news information in the information collection, and determine the event name of the news event corresponding to the information collection according to the multiple words obtained by word segmentation .

According to the multiple words obtained by word segmentation, determine the event name of the news event corresponding to the information set, which can be: remove the semantically repeated words from the multiple words obtained by word segmentation, and determine the part of speech of the remaining words, including nouns and verbs , pronouns, prepositions, etc., connect each word according to the part-of-speech of the remaining words and the preset part-of-speech connection rules to obtain at least one phrase, according to the similarity between each phrase and each phrase in the first preset phrase library , selecting the event name of the news event corresponding to the information set from the at least one phrase.

Wherein, each phrase in the first preset phrase storehouse is the event title of the historical news event of artificial collection in advance. The reason for calculating the similarity between each phrase and the phrases in the first preset phrase library here is to consider the similarity between the current news event and the historical news event, and filter out a phrase with a more fluent and reasonable language logic according to the similarity as Event name.

For example, for a certain information collection, after word segmentation processing is carried out on the information titles of each news information in the information collection, each word is obtained as: Eleventh, National Day, military parade, weapon, introduction. First remove the words with repeated semantics, such as deleting "11", and then determine the part of speech of the remaining words, and connect the words according to the part of speech of the remaining words and the preset part of speech connection rules to get "Introduction to National Day Military Parade Weapons" " and "Introduction to the National Day of Arms Parade", and finally, calculate the similarity between each phrase and each phrase in the first preset phrase library, and obtain a maximum similarity of 98%, and the corresponding phrase of 98% is " National Day military parade weapon introduction", then this phrase is used as the event name of the news event corresponding to the information collection.

In this embodiment, the process of generating the event name of the news event corresponding to each information collection can also be realized through the information clustering model. It can be seen that through this embodiment, not only can multiple information sets be obtained by clustering, but also the event name of the news event corresponding to each information set can be generated, so as to achieve the effect of automatically and efficiently mining news events from news information.

In the above-mentioned step S206, through the news topic creation system, according to a plurality of pieces of news information corresponding to each news event, a corresponding news topic is created for each news event. Wherein, the special topic information is selected from a plurality of news information corresponding to the news event reported by the news topic. In this embodiment, a news topic creation system is provided. The news topic creation system runs in the news topic creation device and is used to execute step S206 to create corresponding news topics for each news event according to multiple news information corresponding to each news event.

In step S206, according to a plurality of pieces of news information corresponding to each news event, a corresponding news topic is created for each news event, specifically:

(b1) For each news event, select information that does not record preset sensitive words and preset sensitive words in multiple news information corresponding to the news event, and the information popularity is greater than the preset popularity, and the selected information is used as feature information on the news feature covering the news event;

(b2) For each news event, word segmentation processing is performed on the information titles of each topic information corresponding to the news event, and among the multiple words obtained by word segmentation, select the word frequency-inverse text frequency index TF-IDF value greater than the preset threshold The word that is selected according to the selected word is used to generate the topic title of the news topic for reporting the news event;

(b3) For each news event, based on the topic information of the news topic used to report the news event and the topic name of the news topic used to report the news event, create a news topic for reporting the news event.

In the above-mentioned action (b1), for each news event, first select the news information that does not have a preset sensitive word in the multiple news information corresponding to the news event (that is, the multiple news information in the information collection corresponding to the news event). In this process, the title and text of the news information can be screened to obtain information that neither the title nor the text has the default sensitive word or default sensitive word recorded. Then, further filter the information whose popularity is greater than the preset popularity among the selected news information, and the information obtained this time is the special topic information of the news topic used to report the news event.

This embodiment provides a method for determining the information popularity of news information, which can be implemented by the news topic creation system, specifically, through the news topic creation system, for each news information, determine the source and reprint amount of the news information and reading volume, and determine the popularity of the news information according to the source, reprint volume, and reading volume of the news information. Among them, the news topic creation system can pre-store a formula for calculating information popularity, which can be exemplified as information popularity S=A*a1+B*b1+C*c1, where A represents the score corresponding to the source of each news information, Here you can preset the score of each source of news information, for example, the score of China Business News is 10 points, the score of Central News is 10 points, a1 is the weight coefficient corresponding to the source of the preset news information, and B represents each news information b1 is the weight coefficient corresponding to the reprint amount of the preset news information, C represents the reading amount of each news information, and c1 is the weight coefficient corresponding to the preset news information reading amount. In this embodiment, the news feature creation system can calculate the information popularity of each news information through this formula.

In the above action (b2), for each news event, word segmentation processing is performed on the information titles of each topic information corresponding to the news event, and among the multiple words obtained by word segmentation, TF-IDF (term frequency-inversedocument frequency, word frequency-inverse text frequency index) value is greater than the preset threshold value, according to the selected word to generate the topic name of the news topic used to report the news event.

For example, if there are two news events in total, for one of the news events, word segmentation is performed on the information titles of each topic information corresponding to the news event, and the TF-IDF value of each word obtained by word segmentation is calculated, so that Among the multiple words, select the word whose TF-IDF value is greater than the preset threshold as the target word, and generate the topic name of the news topic used to report the news event according to the target word. Similarly, for another news event, word segmentation is performed on the information titles of each topic information corresponding to the news event, and the TF-IDF value of each word obtained by word segmentation is calculated, so that among the multiple words obtained by word segmentation , selecting a word whose TF-IDF value is greater than a preset threshold as a target word, and generating a topic name of a news topic used to report the news event according to the target word.

According to the target word, the topic name of the news topic used to report the news event is generated, specifically: remove semantically repetitive words from each target word, and determine the part of speech of the remaining words, including nouns, verbs, pronouns, prepositions, etc. , according to the part-of-speech of each remaining word and the preset part-of-speech connection rule, each word is connected to obtain at least one phrase, and according to the similarity between each phrase and each phrase in the second preset phrase library, in the at least one Select the topic name of the news topic used to report the news event in the phrase.

Wherein, each phrase in the second preset phrase storehouse is the topic title of the historical news topic of artificial collection in advance. Here, the reason for calculating the similarity between each phrase and each phrase in the second preset phrase library is to consider that the current news topic is similar to the historical news topic, and to filter out a phrase with a more fluent and reasonable language logic according to the similarity as Topic name. The process of generating topic names is similar to the process of generating event names mentioned above, and will not be repeated here.

In the above action (b3), for each news event, based on the topic information of the news topic used to report the news event and the topic name of the news topic used to report the news event, create a news topic for reporting the news event . Specifically, for each news event, the news topic creation system creates a news report for reporting the news event based on the topic information of the news topic used to report the news event and the topic name of the news topic used to report the news event. Topics, news topics at least include topic names and topic information. The news feature creation system can also display each created news feature for review by editors.

Figure 3a is a schematic diagram of the interface of the news topic creation system provided by an embodiment of the present application. The event name corresponding to each information collection. In this interface, for each news event, there is also a "Create Topic" button. After the user clicks this button, the news topic creation system will automatically create a corresponding news topic for the news event. Of course, in this embodiment, the "create topic" button may not be set, and the news topic creation system automatically creates a corresponding news topic for the news event.

Figure 3b is a schematic diagram of the interface of the news feature creation system provided by another embodiment of the present application. As shown in Figure 3a and Figure 3b, after the user clicks the "create feature" button, the news feature creation system automatically creates a news feature, and the created news The themes are shown in Figure 3b. Fig. 3c is a schematic diagram of the interface of the news topic creation system provided by another embodiment of the present application. As shown in Fig. 3a and Fig. 3c, if the user clicks on the event name of a certain news event in Fig. 3a, the interface jumps to Fig. 3c As shown, multiple news information corresponding to the news event are displayed.

In this embodiment, the news topic creation system can also display the event heat of each news event as shown in Figure 3a, and the calculation method of the event heat of the news event will be expanded later. The reading volume of each news event is also shown in Fig. 3a, which is the sum of the reading volume of each piece of news information corresponding to each news event.

In this embodiment, the news topic creation system can also generate a corresponding topic description for each news topic after each news topic is created, as shown in Figure 3b. The generated topic description can be generated manually, and can also be generated by the news topic creation system according to the information titles of the multiple topic information contained in each news topic. Specifically, for each news topic, the news topic creation system obtains the information titles of the multiple topic information contained in the news topic, performs word segmentation processing on each news topic, obtains multiple words, and generates the news according to the splicing of the multiple words. Thematic description of the topic.

In this embodiment, the news topic creation system can also be shown in Figure 3c, and the interface of Figure 3c also provides a "create topic" button, and the editors click on the event name of a certain news event to jump to the page shown in Figure 3c. After entering the interface, you can click the "Create Topic" button in this interface to automatically create a news topic for the news event through the news topic creation system.

Further, the method in this embodiment also includes:

(c1) For each news event, determine the source, reprint volume and reading volume of each piece of news information corresponding to the news event, and determine the news information according to the source, reprint volume, and reading volume of each piece of news information corresponding to the news event. Event heat of the event;

(c2) Generate an event popularity ranking list according to the event popularity of each news event, and send the event popularity ranking list to the information provider, so that the information provider can provide news information according to the event popularity ranking list.

In the above action (c1), the event popularity of the news event is determined according to the source, reprint volume, and reading volume of each piece of news information corresponding to the news event, specifically: according to the source of each piece of news information corresponding to the news event, The number of reprints and readings determines the information popularity of each news article, and the sum of the information popularity of each news information is used as the event popularity of the news event. Wherein, the manner of calculating the information popularity of each piece of news information can be as described above, and will not be repeated here. In another embodiment, the sum of the reprints and readings of each piece of news information corresponding to the news event can also be used as the event popularity of the news event.

In the above action (c2), an event popularity ranking is generated according to the event popularity of each news event, and the event popularity ranking is sent to the information provider, so that the information provider can understand the current popularity of each news event and provide hot information , to avoid wasting labor costs by providing non-hot information. The event popularity ranking list can be shown in Table 1 below.

Table 1

event number event name Event heat 1 11 military parade 100 2 so-and-so exercise 98 … … … 10 XX TV series boot 50

Fig. 4 is a schematic diagram of the interface of the user terminal provided by an embodiment of the present application. As shown in Fig. 4, the user terminal can receive and display the news topic pushed by the news topic creation device. As shown in Figure 4, the topic name of the news topic "iphone purchase intention plummets" is displayed in this interface, and the information icons of each topic information are displayed. In this embodiment, when the news topic creation device stores the created news topic, it can be stored in the Hbase database, thereby improving the stability of data storage and the convenience of data calling.

In summary, through the embodiments of the present application, news topics can be automatically created through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency of manually creating news topics and improving the efficiency of creating news topics. Moreover, since the event popularity ranking list can be generated according to the event popularity of each news event, and the event popularity ranking list is sent to the information provider, it is convenient for the information provider to grasp the trend of the current hot event, and it is convenient for the information provider to provide information about the hot event. relevant news.

Fig. 5 is a schematic flow diagram of a method for creating news topics provided by another embodiment of the present application. As shown in Fig. 5, the flow includes the following steps:

Step S502, obtaining a plurality of news information provided by the information provider, the news content corresponding to the news event is recorded in the news information;

Step S504, clustering the multiple pieces of news information through the pre-trained information clustering model to obtain multiple information collections; wherein, the news content of the multiple news information included in each information collection corresponds to the same news event, and different information The news events corresponding to the sets are different;

Step S506, according to a plurality of pieces of news information corresponding to each news event, determine the event heat of each news event, according to the event heat of each news event, select the target event in each news event;

Step S508, create a news topic for the target event through the news topic creation system according to multiple news information corresponding to the target event; wherein, the news topic is used to report the target event; the news topic includes the topic name and topic information, and the topic information is based on the target event The corresponding multiple news articles are determined.

Through the embodiment of this application, it is possible to automatically select a target event with high popularity among various news events, and automatically create a news topic corresponding to the target event through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency in manually creating news topics problems, realize the automatic creation of popular topics, and improve the efficiency of creating news topics.

The specific process of the above step S502 and step S504 is the same as the specific process of the aforementioned step S202 and step S204, and will not be repeated here.

In the above step S506, according to the multiple pieces of news information corresponding to each news event, the event popularity of each news event is determined, specifically: for each news event, the source, reprint amount and reading amount of each news information corresponding to the news event are determined. According to the source, reprint volume and reading volume of each news information corresponding to the news event, the event popularity of the news event is determined. For this process, reference may be made to the description of the aforementioned action (c1), which will not be repeated here. In this embodiment, the source of the news information refers to the information provider of the news information.

In the above step S506, according to the event popularity of each news event, select the target event in each news event, specifically, select the event in each news event whose event popularity is greater than the preset popularity threshold as the target event.

In the above step S508, a news topic is created for the target event through the news topic creation system according to multiple pieces of news information corresponding to the target event. selected from. It can be understood that, according to the foregoing introduction, a news topic may also include a topic description.

The specific process of step S508 is:

(d1) Select information that does not record preset sensitive words and preset sensitive words in multiple news information corresponding to the target event, and the information popularity is greater than the preset popularity, and use the selected information as the special topic information of the news topic;

(d2) Carry out word segmentation processing on the information titles of each topic information, and select words whose word frequency-inverse text frequency index TF-IDF value is greater than the preset threshold value among the multiple words obtained by word segmentation, and generate news topics according to the selected words title of the topic;

(d3) Create a news topic based on the topic name of the news topic and the topic information of the news topic.

For this process, reference may be made to the specific introduction of the aforementioned step S206, which will not be repeated here.

The method in FIG. 5 may also include: generating an event popularity ranking list according to the event popularity of each news event, and sending the event popularity ranking list to the information provider, so that the information provider provides news information according to the event popularity ranking list. Specifically, according to the event popularity of each news event, an event popularity ranking list is generated, and the event popularity ranking list is sent to the information provider, so that the information provider can understand the current popularity of each news event, thereby providing hot information and avoiding providing unreasonable information. Popular information wastes labor costs.

In summary, through the embodiment of this application, it is possible to automatically select a target event with high popularity among various news events, and automatically create a news topic corresponding to the target event through the information clustering model and the news topic creation system, thereby solving the problem of manually creating news. To solve the problem of low efficiency of topics, realize the automatic creation of popular topics and improve the efficiency of creating news topics. Moreover, since the event popularity rankings can be generated according to the event popularity of each news event, and the event popularity rankings are sent to the information provider, it is convenient for the information provider to grasp the trend of the current popular events, and it is convenient for the information provider to provide information related to the popular events. news information.

Fig. 6 is a schematic diagram of the module composition of the news topic creation device provided by an embodiment of the present application. As shown in Fig. 6, the device includes:

The first obtaining module 61 is used to obtain a plurality of pieces of news information provided by the information provider, and the news content corresponding to the news event is recorded in the news information;

The first clustering module 62 is configured to cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information collections; wherein, each of the information collections includes the news of the multiple pieces of news information The content corresponds to the same news event, and the news events corresponding to different information sets are different;

The first creation module 63 is used to create corresponding news topics for each of the news events according to the multiple pieces of news information corresponding to each of the news events through the news topic creation system; wherein, the news topics are used to report the corresponding The news event; the news topic includes a topic name and topic information, and the topic information is determined according to a plurality of pieces of news information corresponding to the news event reported by the news topic.

Optionally, the first clustering module 62 is specifically used for:

Through the pre-trained information clustering model, extract the information titles of the news information described in each article, and calculate the semantic similarity between the information titles of the news information described in each article;

Through the pre-trained information clustering model, according to the semantic similarity between the information titles of the news information described in each article, the news information described in each article is clustered to obtain multiple information sets.

Optionally, also includes a name generation module for:

For each information collection, word segmentation is performed on the information titles of the news articles in the information collection, and the event name of the news event corresponding to the information collection is determined according to the multiple words obtained by word segmentation.

Optionally, the first creation module 63 is specifically used for:

For each of the above-mentioned news events, select the information that does not record the preset sensitive words and preset sensitive words in the multiple news information corresponding to the news event, and the information popularity is greater than the preset popularity, and the selected information is used as the Feature information of the news feature covering the news event;

For each of the news events, word segmentation is performed on the information titles of each topic information corresponding to the news event, and among the multiple words obtained by word segmentation, select the word frequency-inverse text frequency index TF-IDF value greater than the preset threshold Words, generate the topic titles of the news topics used to report the news event according to the selected words;

For each news event, a news topic for reporting the news event is created based on the topic information of the news topic for reporting the news event and the topic name of the news topic for reporting the news event.

Optionally, a thermal determination module is also included for:

Through the news topic creation system, for each news information, determine the source, reprint volume and reading volume of the news information, and determine the information popularity of the news information according to the source, reprint volume and reading volume of the news information.

Optionally, a first sorting module is also included for:

For each said news event, determine the source, reprint volume, and reading volume of each piece of news information corresponding to the news event, and determine the news event based on the source, reprint volume, and reading volume of each piece of news information corresponding to the news event the heat of the event;

Generate an event popularity ranking list according to the event popularity of each of the news events, and send the event popularity ranking list to the information provider for the information provider to provide news information according to the event popularity ranking list.

In the embodiment of the present application, first obtain multiple pieces of news information provided by the information provider, and then cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information sets, each information set includes multiple The news content of each piece of news information corresponds to the same news event, and finally, through the news topic creation system, a corresponding news topic is created for each news event according to the multiple news pieces corresponding to each news event. It can be seen that through the embodiment of the present application, news topics can be automatically created through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency of manually creating news topics and improving the efficiency of creating news topics.

It should be noted that the news topic creation device in the embodiment of the present application can realize the various processes of the aforementioned news topic creation method, and achieve the same effect and function, and will not repeat them here.

Fig. 7 is a schematic diagram of the module composition of the news topic creation device provided by another embodiment of the present application. As shown in Fig. 7, the device includes:

The second obtaining module 71 is used to obtain a plurality of news information provided by the information provider, and the news content corresponding to the news event is recorded in the news information;

The second clustering module 72 is used to cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information collections; wherein, each of the information collections includes the news of multiple pieces of news information The content corresponds to the same news event, and the news events corresponding to different information sets are different;

Selection module 73, is used for determining the event heat of each described news event according to a plurality of pieces of news information corresponding to each described news event, and selects a target event in each of the described news events according to the event heat of each described news event;

The second creation module 74 is used to create a news topic for the target event according to multiple pieces of news information corresponding to the target event through the news topic creation system; wherein, the news topic is used to report the target event; The news topic includes a topic name and topic information, and the topic information is determined according to multiple pieces of news information corresponding to the target event.

Optionally, the second creation module 74 is specifically configured to:

Select information that does not record preset sensitive words and preset sensitive words, and information heat is greater than preset heat among multiple news information corresponding to the target event, and use the selected information as the special topic information of the news topic;

Perform word segmentation processing on the information titles of the topical information described in each article, select words whose word frequency-inverse text frequency index TF-IDF value is greater than the preset threshold value among the multiple words obtained by word segmentation, and generate the news according to the selected words The topic title of the topic;

Create the news topic based on the topic name of the news topic and the topic information of the news topic.

Optionally, the selecting module 73 is specifically used for:

For each news event, determine the source, reprint volume and reading volume of each piece of news information corresponding to the news event;

According to the source, reprint volume and reading volume of each piece of news information corresponding to the news event, the event popularity of the news event is determined.

Optionally, a second sorting module is also included for:

Generate an event popularity ranking list according to the event popularity of each of the news events, and send the event popularity ranking list to the information provider for the information provider to provide news information according to the event popularity ranking list.

Through the embodiment of this application, it is possible to automatically select a target event with high popularity among various news events, and automatically create a news topic corresponding to the target event through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency in manually creating news topics problems, realize the automatic creation of popular topics, and improve the efficiency of creating news topics.

It should be noted that the news topic creation device in the embodiment of the present application can realize the various processes of the aforementioned news topic creation method, and achieve the same effect and function, and will not repeat them here.

The embodiment of the present application also provides a news feature creation device. Figure 8 is a schematic structural diagram of the news feature creation device provided in an embodiment of the present application. A relatively large difference may include one or more processors 901 and memory 902, and one or more storage application programs or data may be stored in the memory 902. Wherein, the storage 902 may be a short-term storage or a persistent storage. The application program stored in the memory 902 may include one or more modules (not shown in the figure), and each module may include a series of computer-executable instructions to the news feature creation device. Furthermore, the processor 901 can be configured to communicate with the memory 902, and execute a series of computer-executable instructions in the memory 902 on the news feature creation device. The news feature creation device may also include one or more power supplies 903, one or more wired or wireless network interfaces 904, one or more input and output interfaces 905, one or more keyboards 906, and the like.

In a specific embodiment, the news feature creation device includes a memory, and one or more programs, wherein one or more programs are stored in the memory, and one or more programs may include one or more modules, and Each module may include a series of computer-executable instructions in the news topic creation device, and the one or more programs configured to be executed by one or more processors include computer-executable instructions for performing the following:

Obtain multiple news information provided by the information provider, and the news information corresponding to the news event is recorded in the news information;

The multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets; wherein, the news content of the multiple pieces of news information included in each information set corresponds to the same news event, and different The news events corresponding to the above information sets are different;

Through the news topic creation system, according to the multiple news information corresponding to each of the news events, a corresponding news topic is created for each of the news events; wherein, the news topic is used to report the corresponding news event; the news A topic includes a topic name and topic information, and the topic information is determined based on a plurality of pieces of news information corresponding to news events reported in the news topic.

Optionally, when the computer-executable instructions are executed, the multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets, including:

Through the pre-trained information clustering model, extract the information titles of the news information described in each article, and calculate the semantic similarity between the information titles of the news information described in each article;

Through the pre-trained information clustering model, according to the semantic similarity between the information titles of the news information described in each article, the news information described in each article is clustered to obtain multiple information sets.

Optionally, when executed, the computer-executable instructions further include:

For each information collection, word segmentation is performed on the information titles of the news articles in the information collection, and the event name of the news event corresponding to the information collection is determined according to the multiple words obtained by word segmentation.

Optionally, when the computer-executable instructions are executed, according to the multiple pieces of news information corresponding to each of the news events, a corresponding news topic is created for each of the news events, including:

For each of the above-mentioned news events, select the information that does not record the preset sensitive words and preset sensitive words in the multiple news information corresponding to the news event, and the information popularity is greater than the preset popularity, and the selected information is used as the Feature information of the news feature covering the news event;

For each of the news events, word segmentation is performed on the information titles of each topic information corresponding to the news event, and among the multiple words obtained by word segmentation, select the word frequency-inverse text frequency index TF-IDF value greater than the preset threshold Words, generate the topic titles of the news topics used to report the news event according to the selected words;

For each news event, a news topic for reporting the news event is created based on the topic information of the news topic for reporting the news event and the topic name of the news topic for reporting the news event.

Optionally, when executed, the computer-executable instructions further include:

Through the news topic creation system, for each news information, determine the source, reprint volume and reading volume of the news information, and determine the information popularity of the news information according to the source, reprint volume and reading volume of the news information.

Optionally, when executed, the computer-executable instructions further include:

For each said news event, determine the source, reprint volume, and reading volume of each piece of news information corresponding to the news event, and determine the news event based on the source, reprint volume, and reading volume of each piece of news information corresponding to the news event the heat of the event;

Generate an event popularity ranking list according to the event popularity of each of the news events, and send the event popularity ranking list to the information provider for the information provider to provide news information according to the event popularity ranking list.

In the embodiment of the present application, first obtain multiple pieces of news information provided by the information provider, and then cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information sets, each information set includes multiple The news content of each piece of news information corresponds to the same news event, and finally, through the news topic creation system, a corresponding news topic is created for each news event according to the multiple news pieces corresponding to each news event. It can be seen that through the embodiment of the present application, news topics can be automatically created through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency of manually creating news topics and improving the efficiency of creating news topics.

It should be noted that the news topic creation device in the embodiment of the present application can realize the various processes of the aforementioned news topic creation method, and achieve the same effect and function, and will not repeat them here.

In another specific embodiment, the news feature creation device includes a memory, and one or more programs, wherein one or more programs are stored in the memory, and one or more programs may include one or more modules, And each module may include a series of computer-executable instructions in the news topic creation device, and the one or more programs configured to be executed by one or more processors include computer-executable instructions for performing the following:

Obtain multiple news information provided by the information provider, and the news information corresponding to the news event is recorded in the news information;

The multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets; wherein, the news content of the multiple pieces of news information included in each information set corresponds to the same news event, and different The news events corresponding to the above information sets are different;

According to a plurality of pieces of news information corresponding to each of the news events, determine the event heat of each of the news events, and select the target event in each of the news events according to the event heat of each of the news events;

Create a news topic for the target event through the news topic creation system according to multiple news information corresponding to the target event; wherein, the news topic is used to report the target event; the news topic includes topic name and topic Information, the topical information is determined according to multiple pieces of news information corresponding to the target event.

Optionally, when the computer-executable instructions are executed, according to multiple pieces of news information corresponding to the target event, a news topic is created for the target event, including:

Select information that does not record preset sensitive words and preset sensitive words, and information heat is greater than preset heat among multiple news information corresponding to the target event, and use the selected information as the special topic information of the news topic;

Perform word segmentation processing on the information titles of the topical information described in each article, select words whose word frequency-inverse text frequency index TF-IDF value is greater than the preset threshold value among the multiple words obtained by word segmentation, and generate the news according to the selected words The topic title of the topic;

Create the news topic based on the topic name of the news topic and the topic information of the news topic.

Optionally, when the computer-executable instructions are executed, the event popularity of each of the news events is determined according to the plurality of news information corresponding to each of the news events, including:

For each news event, determine the source, reprint volume and reading volume of each piece of news information corresponding to the news event;

According to the source, reprint volume and reading volume of each piece of news information corresponding to the news event, the event popularity of the news event is determined.

Optionally, when executed, the computer-executable instructions further include:

Generate an event popularity ranking list according to the event popularity of each of the news events, and send the event popularity ranking list to the information provider for the information provider to provide news information according to the event popularity ranking list.

Through the embodiment of this application, it is possible to automatically select a target event with high popularity among various news events, and automatically create a news topic corresponding to the target event through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency in manually creating news topics problems, realize the automatic creation of popular topics, and improve the efficiency of creating news topics.

It should be noted that the news topic creation device in the embodiment of the present application can realize the various processes of the aforementioned news topic creation method, and achieve the same effect and function, and will not repeat them here.

Further, the embodiment of the present application also provides a storage medium for storing computer-executable instructions. In a specific embodiment, the storage medium may be a USB flash drive, an optical disk, a hard disk, etc. When executable instructions are executed by the processor, the following processes can be realized:

Obtain multiple news information provided by the information provider, and the news information corresponding to the news event is recorded in the news information;

The multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets; wherein, the news content of the multiple pieces of news information included in each information set corresponds to the same news event, and different The news events corresponding to the above information sets are different;

Through the news topic creation system, according to the multiple news information corresponding to each of the news events, a corresponding news topic is created for each of the news events; wherein, the news topic is used to report the corresponding news event; the news A topic includes a topic name and topic information, and the topic information is determined based on a plurality of pieces of news information corresponding to news events reported in the news topic.

Optionally, when the computer-executable instructions stored in the storage medium are executed by the processor, the multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets, including:

Through the pre-trained information clustering model, extract the information titles of the news information described in each article, and calculate the semantic similarity between the information titles of the news information described in each article;

Through the pre-trained information clustering model, according to the semantic similarity between the information titles of the news information described in each article, the news information described in each article is clustered to obtain multiple information sets.

Optionally, when executed by the processor, the computer-executable instructions stored in the storage medium further include:

For each information collection, word segmentation is performed on the information titles of the news articles in the information collection, and the event name of the news event corresponding to the information collection is determined according to the multiple words obtained by word segmentation.

Optionally, when the computer-executable instructions stored in the storage medium are executed by the processor, a corresponding news topic is created for each of the news events according to multiple pieces of news information corresponding to each of the news events, including:

For each of the above-mentioned news events, select the information that does not record the preset sensitive words and preset sensitive words in the multiple news information corresponding to the news event, and the information popularity is greater than the preset popularity, and the selected information is used as the Feature information of the news feature covering the news event;

For each of the news events, word segmentation is performed on the information titles of each topic information corresponding to the news event, and among the multiple words obtained by word segmentation, select the word frequency-inverse text frequency index TF-IDF value greater than the preset threshold Words, generate the topic titles of the news topics used to report the news event according to the selected words;

For each news event, a news topic for reporting the news event is created based on the topic information of the news topic for reporting the news event and the topic name of the news topic for reporting the news event.

Optionally, when executed by the processor, the computer-executable instructions stored in the storage medium further include:

Through the news topic creation system, for each news information, determine the source, reprint volume and reading volume of the news information, and determine the information popularity of the news information according to the source, reprint volume and reading volume of the news information.

Optionally, when executed by the processor, the computer-executable instructions stored in the storage medium further include:

For each said news event, determine the source, reprint volume, and reading volume of each piece of news information corresponding to the news event, and determine the news event based on the source, reprint volume, and reading volume of each piece of news information corresponding to the news event the heat of the event;

Generate an event popularity ranking list according to the event popularity of each of the news events, and send the event popularity ranking list to the information provider for the information provider to provide news information according to the event popularity ranking list.

In the embodiment of the present application, first obtain multiple pieces of news information provided by the information provider, and then cluster the multiple pieces of news information through a pre-trained information clustering model to obtain multiple information sets, each information set includes multiple The news content of each piece of news information corresponds to the same news event, and finally, through the news topic creation system, a corresponding news topic is created for each news event according to the multiple news pieces corresponding to each news event. It can be seen that through the embodiment of the present application, news topics can be automatically created through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency of manually creating news topics and improving the efficiency of creating news topics.

It should be noted that the storage medium in the embodiment of the present application can realize the various processes of the aforementioned method for creating a news topic, and achieve the same effect and function, which will not be repeated here.

In another specific embodiment, the storage medium can be a USB flash drive, an optical disc, a hard disk, etc., and the computer-executable instructions stored in the storage medium can realize the following process when executed by the processor:

Obtain multiple news information provided by the information provider, and the news information corresponding to the news event is recorded in the news information;

The multiple pieces of news information are clustered through a pre-trained information clustering model to obtain multiple information sets; wherein, the news content of the multiple pieces of news information included in each information set corresponds to the same news event, and different The news events corresponding to the above information sets are different;

According to a plurality of pieces of news information corresponding to each of the news events, determine the event heat of each of the news events, and select the target event in each of the news events according to the event heat of each of the news events;

Create a news topic for the target event through the news topic creation system according to multiple news information corresponding to the target event; wherein, the news topic is used to report the target event; the news topic includes topic name and topic Information, the topical information is determined according to multiple pieces of news information corresponding to the target event.

Optionally, when the computer-executable instructions stored in the storage medium are executed by the processor, a news topic is created for the target event according to multiple news articles corresponding to the target event, including:

Select information that does not record preset sensitive words and preset sensitive words, and information heat is greater than preset heat among multiple news information corresponding to the target event, and use the selected information as the special topic information of the news topic;

Perform word segmentation processing on the information titles of the topical information described in each article, select words whose word frequency-inverse text frequency index TF-IDF value is greater than the preset threshold value among the multiple words obtained by word segmentation, and generate the news according to the selected words The topic title of the topic;

Create the news topic based on the topic name of the news topic and the topic information of the news topic.

Optionally, when the computer-executable instructions stored in the storage medium are executed by the processor, the event popularity of each of the news events is determined according to the plurality of news information corresponding to each of the news events, including:

For each news event, determine the source, reprint volume and reading volume of each piece of news information corresponding to the news event;

According to the source, reprint volume and reading volume of each piece of news information corresponding to the news event, the event popularity of the news event is determined.

Optionally, when executed by the processor, the computer-executable instructions stored in the storage medium further include:

Generate an event popularity ranking list according to the event popularity of each of the news events, and send the event popularity ranking list to the information provider for the information provider to provide news information according to the event popularity ranking list.

Through the embodiment of this application, it is possible to automatically select a target event with high popularity among various news events, and automatically create a news topic corresponding to the target event through the information clustering model and the news topic creation system, thereby solving the problem of low efficiency in manually creating news topics problems, realize the automatic creation of popular topics, and improve the efficiency of creating news topics.

It should be noted that the storage medium in the embodiment of the present application can realize the various processes of the aforementioned method for creating a news topic, and achieve the same effect and function, which will not be repeated here.

In the 1990s, the improvement of a technology can be clearly distinguished as an improvement in hardware (for example, improvements in circuit structures such as diodes, transistors, and switches) or improvements in software (improvement in method flow). However, with the development of technology, the improvement of many current method flows can be regarded as the direct improvement of the hardware circuit structure. Designers almost always get the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement of a method flow cannot be realized with a hardware physical module. For example, a Programmable Logic Device (Programmable Logic Device, PLD) (such as a Field Programmable Gate Array (Field Programmable Gate Array, FPGA)) is such an integrated circuit, and its logic function is determined by programming the device by a user. A digital system is "integrated" on a PLD by the designer's own programming, without the need for a chip manufacturer to design and manufacture a dedicated integrated circuit chip. Moreover, nowadays, instead of making integrated circuit chips by hand, this kind of programming is mostly realized by "logic compiler (logic compiler)" software, which is similar to the software compiler used when writing programs. The original code of the computer must also be written in a specific programming language, which is called a hardware description language (Hardware Description Language, HDL), and there is not only one kind of HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language) , AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., currently the most commonly used is VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. It should also be clear to those skilled in the art that only a little logic programming of the method flow in the above-mentioned hardware description languages and programming into an integrated circuit can easily obtain a hardware circuit for realizing the logic method flow.

The controller may be implemented in any suitable way, for example the controller may take the form of a microprocessor or processor and a computer readable medium storing computer readable program code (such as software or firmware) executable by the (micro)processor , logic gates, switches, Application Specific Integrated Circuit (ASIC), programmable logic controllers, and embedded microcontrollers, examples of controllers include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the memory's control logic. Those skilled in the art also know that, in addition to realizing the controller in a purely computer-readable program code mode, it is entirely possible to make the controller use logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded The same function can be realized in the form of a microcontroller or the like. Therefore, this kind of controller can be considered as a hardware component, and the devices included in it for realizing various functions can also be regarded as the structure in the hardware component. Or even, the means for implementing various functions can be regarded as a structure within both a software module implementing a method and a hardware component.

The systems, devices, modules or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementing device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or Combinations of any of these devices.

For the convenience of description, when describing the above devices, functions are divided into various units and described separately. Of course, when implementing the present application, the functions of each unit can be implemented in one or more pieces of software and/or hardware.

Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each process and/or block in the flowchart and/or block diagram, and a combination of processes and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

Memory may include non-permanent storage in computer-readable media, in the form of random access memory (RAM) and/or nonvolatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer readable media.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may be implemented by any method or technology for storage of information. Information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for computers include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Flash memory or other memory technology, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cartridge, tape magnetic disk storage or other magnetic storage device or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media excludes transitory computer-readable media, such as modulated data signals and carrier waves.

It should also be noted that the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes Other elements not expressly listed, or elements inherent in the process, method, commodity, or apparatus are also included. Without further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems or computer program products. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

This application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including storage devices.

Each embodiment in this specification is described in a progressive manner, the same and similar parts of each embodiment can be referred to each other, and each embodiment focuses on the difference from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for relevant parts, please refer to the part of the description of the method embodiment.

The above descriptions are only examples of the present application, and are not intended to limit the present application. For those skilled in the art, various modifications and changes may occur in this application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the scope of the claims of the present application.

Claims (20)

1. A news topic creation method comprising:

acquiring a plurality of pieces of news information provided by an information provider, wherein news contents corresponding to news events are recorded in the news information;

clustering the news information through a pre-trained information clustering model to obtain a plurality of information sets; the news content of a plurality of pieces of news information contained in each information set corresponds to the same news event, and the news events corresponding to different information sets are different;

Extracting thematic information and thematic names according to a plurality of pieces of news information corresponding to each news event through a news thematic creation system, and creating corresponding news thematic for each news event based on the extracted thematic information and thematic names; the news topics are used for reporting the corresponding news events; the news topics comprise topic names and topic information, and the topic information is obtained by determining a plurality of pieces of news information corresponding to news events reported by the news topics.

2. The method of claim 1, clustering the plurality of news information by a pre-trained information clustering model to obtain a plurality of information sets, comprising:

extracting information titles of all the news information through a pre-trained information clustering model, and calculating semantic similarity among the information titles of all the news information;

clustering each piece of news information according to semantic similarity among information titles of each piece of news information through a pre-trained information clustering model to obtain a plurality of information sets.

3. The method of claim 2, further comprising:

For each information set, word segmentation is carried out on the information titles of all news information in the information set, and event names of news events corresponding to the information set are determined according to a plurality of words obtained by word segmentation.

4. The method of claim 1, extracting topic information and topic names from a plurality of news information corresponding to each of the news events, creating corresponding news topics for each of the news events based on the extracted topic information and topic names, comprising:

selecting information which is not recorded with preset sensitive words and has information heat greater than preset heat from a plurality of pieces of news information corresponding to each news event, and taking the selected information as thematic information of news themes for reporting the news event;

for each news event, carrying out word segmentation on the information title of each piece of topic information corresponding to the news event, selecting words with word frequency-inverse text frequency index TF-IDF value larger than a preset threshold value from a plurality of words obtained by word segmentation, and generating topic names for reporting news topics of the news event according to the selected words;

For each of the news events, creating a news topic for reporting the news event based on topic information of the news topic for reporting the news event and a topic name of the news topic for reporting the news event.

5. The method of claim 4, further comprising:

and determining the source, the transfer capacity and the reading quantity of the news information according to each piece of news information by a news topic creation system, and determining the information heat of the news information according to the source, the transfer capacity and the reading quantity of the news information.

6. The method of any one of claims 1 to 5, further comprising:

determining the source, the transfer capacity and the reading quantity of each piece of news information corresponding to each news event according to each news event, and determining the event heat of the news event according to the source, the transfer capacity and the reading quantity of each piece of news information corresponding to the news event;

generating an event popularity ranking list according to the event popularity of each news event, and sending the event popularity ranking list to the information provider so that the information provider can provide news information according to the event popularity ranking list.

7. A news topic creation method comprising:

Acquiring a plurality of pieces of news information provided by an information provider, wherein news contents corresponding to news events are recorded in the news information;

clustering the news information through a pre-trained information clustering model to obtain a plurality of information sets; the news content of a plurality of pieces of news information contained in each information set corresponds to the same news event, and the news events corresponding to different information sets are different;

determining the event heat of each news event according to a plurality of pieces of news information corresponding to each news event, and selecting a target event from each news event according to the event heat of each news event;

extracting thematic information and thematic names according to a plurality of pieces of news information corresponding to the target event by a news thematic creation system, and creating news thematic for the target event based on the extracted thematic information and thematic names; wherein the news topic is used for reporting the target event; the news topics comprise topic names and topic information, and the topic information is determined and obtained according to a plurality of pieces of news information corresponding to the target event.

8. The method of claim 7, extracting topic information and topic names from a plurality of news information corresponding to the target event, creating news topics for the target event based on the extracted topic information and topic names, comprising:

selecting information which is not recorded with preset sensitive words and has information heat degree larger than preset heat degree from a plurality of pieces of news information corresponding to the target event, and taking the selected information as thematic information of the news themes;

word segmentation is carried out on the information titles of the thematic information, words with word frequency-inverse text frequency index TF-IDF values larger than a preset threshold value are selected from a plurality of words obtained through word segmentation, and thematic names of the news thematic are generated according to the selected words;

creating the news topic based on the topic name of the news topic and topic information of the news topic.

9. The method of claim 7 or 8, determining an event popularity of each of the news events based on a plurality of news information corresponding to each of the news events, comprising:

determining the source, the reloading amount and the reading amount of each piece of news information corresponding to each news event;

And determining the event heat of the news event according to the source, the transfer capacity and the reading quantity of each piece of news information corresponding to the news event.

10. The method of claim 9, further comprising:

generating an event popularity ranking list according to the event popularity of each news event, and sending the event popularity ranking list to the information provider so that the information provider can provide news information according to the event popularity ranking list.

11. A news topic creation device comprising:

the first acquisition module is used for acquiring a plurality of pieces of news information provided by the information provider, wherein the news information is recorded with news content corresponding to a news event;

the first clustering module is used for clustering the plurality of news information through a pre-trained information clustering model to obtain a plurality of information sets; the news content of a plurality of pieces of news information contained in each information set corresponds to the same news event, and the news events corresponding to different information sets are different;

the first creation module is used for extracting thematic information and thematic names according to a plurality of pieces of news information corresponding to each news event through the news thematic creation system, and creating corresponding news themes for each news event based on the extracted thematic information and the thematic names; the news topics are used for reporting the corresponding news events; the news topics comprise topic names and topic information, and the topic information is obtained by determining a plurality of pieces of news information corresponding to news events reported by the news topics.

12. The apparatus of claim 11, the first clustering module is specifically configured to:

extracting information titles of all the news information through a pre-trained information clustering model, and calculating semantic similarity among the information titles of all the news information;

clustering each piece of news information according to semantic similarity among information titles of each piece of news information through a pre-trained information clustering model to obtain a plurality of information sets.

13. The apparatus of claim 11, the first creation module is specifically configured to:

selecting information which is not recorded with preset sensitive words and has information heat greater than preset heat from a plurality of pieces of news information corresponding to each news event, and taking the selected information as thematic information of news themes for reporting the news event;

for each news event, carrying out word segmentation on the information title of each piece of topic information corresponding to the news event, selecting words with word frequency-inverse text frequency index TF-IDF value larger than a preset threshold value from a plurality of words obtained by word segmentation, and generating topic names for reporting news topics of the news event according to the selected words;

For each of the news events, creating a news topic for reporting the news event based on topic information of the news topic for reporting the news event and a topic name of the news topic for reporting the news event.

14. The apparatus of any of claims 11 to 13, further comprising a first ranking module to:

determining the source, the transfer capacity and the reading quantity of each piece of news information corresponding to each news event according to each news event, and determining the event heat of the news event according to the source, the transfer capacity and the reading quantity of each piece of news information corresponding to the news event;

generating an event popularity ranking list according to the event popularity of each news event, and sending the event popularity ranking list to the information provider so that the information provider can provide news information according to the event popularity ranking list.

15. A news topic creation device comprising:

the second acquisition module is used for acquiring a plurality of pieces of news information provided by the information provider, wherein the news information is recorded with news content corresponding to a news event;

the second clustering module is used for clustering the plurality of news information through a pre-trained information clustering model to obtain a plurality of information sets; the news content of a plurality of pieces of news information contained in each information set corresponds to the same news event, and the news events corresponding to different information sets are different;

The selecting module is used for determining the event heat of each news event according to a plurality of pieces of news information corresponding to each news event, and selecting a target event from each news event according to the event heat of each news event;

the second creation module is used for extracting topic information and topic names according to a plurality of pieces of news information corresponding to the target event through a news topic creation system, and creating news topics for the target event based on the extracted topic information and topic names; wherein the news topic is used for reporting the target event; the news topics comprise topic names and topic information, and the topic information is determined and obtained according to a plurality of pieces of news information corresponding to the target event.

16. The apparatus of claim 15, the second creation module is specifically configured to:

selecting information which is not recorded with preset sensitive words and has information heat degree larger than preset heat degree from a plurality of pieces of news information corresponding to the target event, and taking the selected information as thematic information of the news themes;

word segmentation is carried out on the information titles of the thematic information, words with word frequency-inverse text frequency index TF-IDF values larger than a preset threshold value are selected from a plurality of words obtained through word segmentation, and thematic names of the news thematic are generated according to the selected words;

Creating the news topic based on the topic name of the news topic and topic information of the news topic.

17. The apparatus according to claim 15 or 16, wherein the selection module is specifically configured to:

determining the source, the reloading amount and the reading amount of each piece of news information corresponding to each news event;

and determining the event heat of the news event according to the source, the transfer capacity and the reading quantity of each piece of news information corresponding to the news event.

18. The apparatus of claim 17, further comprising a second ranking module to:

generating an event popularity ranking list according to the event popularity of each news event, and sending the event popularity ranking list to the information provider so that the information provider can provide news information according to the event popularity ranking list.

19. A news topic creation device comprising: a processor; and a memory arranged to store computer executable instructions that when executed cause the processor to implement the steps of the news topic creation method of any one of the preceding claims 1 to 6 or the news topic creation method of any one of the preceding claims 7 to 10.

20. A storage medium storing computer-executable instructions which, when executed, implement the steps of the news topic creation method of any one of the preceding claims 1 to 6 or the steps of the news topic creation method of any one of the preceding claims 7 to 10.