CN106954087B - method and device for recommending advertisement short films and video programs in mixed mode - Google Patents

Abstract

The invention provides a method and a device for recommending advertisement short films and video programs in a mixed mode. The method is applied to a server which is in communication connection with the user terminal. The server comprises a database, wherein the database comprises a first vector for describing the interest degree of the user in the advertisement short film and a second vector for describing the interest degree of the user in the video program. Firstly, acquiring a first vector and a second vector; then selecting vector elements meeting preset conditions from the first vector and the second vector according to preset conditions, and processing the selected vector elements to obtain a mixed recommended vector; and further obtaining a mixed recommendation list of the advertisement short films and the video programs, and playing the advertisement short films and the video programs by the user terminal according to the mixed recommendation list. Therefore, the advertisement short film is not inserted into the video program for delivery any more in the advertisement short film delivery mode, but the advertisement short film is also regarded as the video program, and meanwhile, the user experience is improved.

Description

Advertisement short film and video program hybrid recommendation method and device

technical field

The invention relates to the field of advertisement delivery, in particular to a method and device for recommending a mixture of advertisement clips and video programs.

Background technique

Existing short commercials are basically based on inserting advertisements through pre-rolls during the playing of video programs. This form of delivery (for example, pre-insert, mid-insert advertisements, etc.) is relatively simple, and causes users to be harassed by advertisements in the process of watching video programs, which affects the user's viewing experience of video programs.

Contents of the invention

In view of this, the present invention provides a mixed recommendation method and device for short commercials and video programs, which can treat short commercials and video programs equally, and generate mixed recommendations for short commercials and video programs according to the user's interest in short commercials and video programs list to improve user experience.

A preferred embodiment of the present invention provides a method for recommending a mixture of advertising clips and video programs, which is applied to a server connected to a user terminal through communication. A vector, a second vector used to describe the degree of interest in the video program, said method comprising:

obtaining the first vector and the second vector;

Selecting vector elements satisfying preset conditions from the first vector and the second vector respectively, and processing the selected vector elements to form a mixed recommendation vector;

A mixed recommendation list of commercial short films and video programs is obtained according to the mixed recommendation vectors, so that the user terminal plays commercial short films and video programs based on the mixed recommendation list.

A preferred embodiment of the present invention also provides a device for recommending mixed advertisement clips and video programs, which is applied to a server communicatively connected to a user terminal. The server includes a database, and the database stores information used to describe the user's degree of interest in advertisement clips. The first vector, the second vector used to describe the degree of interest in the video program, the device includes:

An acquisition module, configured to acquire the first vector and the second vector;

A mixed recommendation vector module, configured to select vector elements satisfying preset conditions from the first vector and the second vector, and process the selected vector elements to form a mixed recommendation vector;

The recommendation list generating module is configured to obtain a mixed recommendation list of commercial short films and video programs according to the mixed recommendation vector, so that the user terminal can play commercial short films and video programs based on the mixed recommendation list.

Compared with the prior art, the present invention has the following beneficial effects:

The present invention provides a method and device for recommending mixed advertisement clips and video programs, which are applied to a server communicating with a user terminal. Wherein, the server includes a database storing the first vector and the second vector. The first vector is used to describe the user's degree of interest in the advertisement clip, and the second vector is used to describe the user's degree of interest in the video program. Select vector elements satisfying preset conditions from the obtained first vector and second vector, and process the selected vector elements to obtain a mixed recommendation vector, so as to obtain a mixed recommendation list of advertisement clips and video programs. In this way, the user terminal plays the short commercial and the video program according to the mixed recommendation list, so that the user will not be disturbed by the short commercial while watching the video program, thereby improving user experience.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and thus It should be regarded as a limitation on the scope, and those skilled in the art can also obtain other related drawings based on these drawings without creative work.

Fig. 1 is a schematic diagram of communication between a server and at least one user terminal provided by a preferred embodiment of the present invention.

FIG. 2 is a schematic block diagram of the server shown in FIG. 1 .

Fig. 3 is a schematic flow chart of a mixed recommendation method for commercial shorts and video programs provided by a preferred embodiment of the present invention.

FIG. 4 is a schematic flowchart of the sub-steps included in step S130 in FIG. 3 .

FIG. 5 is a schematic flowchart of substeps included in substep S132 in FIG. 4 .

Fig. 6 is a schematic diagram of obtaining a mixed recommendation vector provided by a preferred embodiment of the present invention.

FIG. 7 is a schematic flowchart of the sub-steps included in step S140 in FIG. 3 .

Fig. 8 is another schematic flow chart of a method for recommending a mixture of advertising clips and video programs provided by a preferred embodiment of the present invention.

FIG. 9 is a schematic flowchart of some sub-steps included in step S110 in FIG. 8 .

FIG. 10 is a schematic flowchart of another part of sub-steps included in step S110 in FIG. 8 .

Fig. 11 is a schematic block diagram of a device for recommending mixed advertisement clips and video programs according to a preferred embodiment of the present invention.

Fig. 12 is another schematic block diagram of a device for recommending mixed advertisement clips and video programs according to a preferred embodiment of the present invention.

Icons: 100-server; 110-storage; 112-database; 120-storage controller; 130-processor; Recommendation vector module; 331-selection sub-module; 332-processing sub-module; 340-recommendation list generation module.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. The components of the embodiments of the invention generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts belong to the protection scope of the present invention.

It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

Please refer to FIG. 1 , which is a schematic diagram of communication between a server 100 and at least one user terminal 200 provided by a preferred embodiment of the present invention. The server 100 communicates with the user terminal 200 through a network, an interface, and the like. Wherein, the user terminal 200 may be, but not limited to, a TV set, a smart phone, a tablet computer, and the like. The user terminal 200 sends a request to the server 100 for obtaining a recommendation list; the server 100 receives the request, and sends a mixed recommendation list of advertisement clips and video programs to the user terminal 200 . The user terminal 200 receives and plays the advertisement clip and video program according to the mixed recommendation list, so that the user can watch the complete video program.

Please refer to FIG. 2 , which is a schematic block diagram of the server 100 shown in FIG. 1 . The server 100 may include a database 112 , a hybrid recommendation device 300 , a storage 110 , a storage controller 120 and a processor 130 .

The components of the memory 110 , the storage controller 120 and the processor 130 are directly or indirectly electrically connected to realize data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. A database 112 and a hybrid recommendation device 300 are stored in the memory 110, and the hybrid recommendation device 300 includes at least one software function module that can be stored in the memory 110 in the form of software or firmware. The processor 130 executes various functional applications and data processing by running software programs and modules stored in the memory 110, such as the hybrid recommendation device 300 in the embodiment of the present invention, that is, realizes the advertising video in the embodiment of the present invention Hybrid recommendation method with video titles.

Wherein, the memory 110 may be, but not limited to, random access memory (Random Access Memory, RAM), read-only memory (Read Only Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), which can Erasable Programmable Read-Only Memory (EPROM), Electric Erasable Programmable Read-Only Memory (EEPROM), etc. Wherein, the memory 110 is used to store a program, and the processor 130 executes the program after receiving an execution instruction. The access of the processor 130 and possibly other components to the memory 110 may be performed under the control of the memory controller 120 .

The processor 130 may be an integrated circuit chip and has a signal processing capability. The aforementioned processor 130 may be a general processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), and the like. It can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. Various methods, steps and logic block diagrams disclosed in the embodiments of the present invention may be implemented or executed. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, and the like.

It can be understood that the structure shown in FIG. 2 is only for illustration, and the server 100 may also include more or less components than those shown in FIG. 2 , or have a configuration different from that shown in FIG. 2 . Each component shown in Fig. 2 may be implemented by hardware, software or a combination thereof.

Please refer to FIG. 3 . FIG. 3 is a schematic flow chart of a method for recommending a mixture of advertising clips and video programs provided by a preferred embodiment of the present invention. The method is applied to a server 100 communicatively connected to a user terminal 200 . The server 100 includes a database 112, and the database 112 stores a first vector used to describe the user's degree of interest in the advertisement clip and a second vector used to describe the user's degree of interest in the video program. The process in FIG. 3 may be implemented by the processor 130 . The specific process of the mixed recommendation method for advertising clips and video programs will be described in detail below.

Step S120, acquiring the first vector and the second vector.

In this embodiment, the database 112 stores information of multiple user terminals 200 . For one of the user terminals 200, information corresponding to the user terminal 200 is extracted. Wherein, the information includes the first vector and the second vector. The first vector includes at least one advertisement clip and the similarity between each advertisement clip and the user using the user terminal 200 . The second vector includes at least one video program and the similarity between each video program and the user using the user terminal 200 . The user's degree of interest in the advertisement clip or video program can be obtained through the similarity. Generally, the higher the similarity, the more interested the user is.

Step S130, selecting vector elements satisfying preset conditions from the first vector and the second vector respectively, and processing the selected vector elements to form a mixed recommendation vector.

Please refer to FIG. 4 , which is a schematic flowchart of the sub-steps included in step S130 in FIG. 3 . The step S130 may include a sub-step S131 and a sub-step S132.

In the sub-step S131, vector elements satisfying preset conditions are respectively selected from the first vector and the second vector to form a third vector and a fourth vector.

In this embodiment, the method of selecting vector elements satisfying preset conditions from the first vector and the second vector to form the third vector and the fourth vector is any one of the following two.

In the first vector, select the first preset number of vector elements in the order of similarity from high to low to form the third vector, and in the second vector, select the second preset number in the order of similarity from high to low It is assumed that a number of vector elements form a fourth vector, wherein the vector elements in the first vector and the second vector are arranged in descending order or ascending order according to the similarity of the vector elements.

Select vector elements whose similarity of vector elements is greater than the first preset similarity in the first vector to form a third vector, and select vector elements whose similarity of vector elements is greater than the second preset similarity in the second vector to form Fourth vector.

In an implementation manner of this embodiment, the vector elements (commercial clips) in the first vector and the vector elements (video programs) in the second vector are arranged in descending or ascending order according to the similarity of the vector elements. When selecting vector elements, select from the first vector element with the highest similarity in descending order of similarity. A first preset number of vector elements is selected from the first vector to form a third vector, and the third vector includes at least one advertisement clip and the similarity between each advertisement clip and the user. The vector elements of the second preset data are selected from the second vector to form a fourth vector, and the fourth vector includes at least one video program and the similarity between each video program and the user. For example, 5 vector elements may be selected in the first vector to recommend 5 commercial clips; 20 vector elements may be selected in the second vector to recommend 20 video programs. As a result, there are theoretical and rule-based controls on the mixed programming of commercial clips and video programs, and a good balance can be achieved in terms of advertising revenue and user experience.

Wherein, the specific value of the first preset number is determined according to the user's degree of interest in short commercials, and the specific value of the second preset number is determined according to the user's interest in video programs. In the case that a user basically does not click on an advertisement or quits after clicking on it, it is possible to recommend fewer short advertisements and reduce the proportion of short advertisements. Thus, according to the interest of the user, the recommended number of commercial videos, the recommended number of video programs, and the ratio of commercial videos to video programs are determined.

In another implementation manner of this embodiment, vector element selection may be performed by setting a preset similarity. Specifically, select vector elements (advertising clips) whose similarity is greater than the first preset similarity in the first vector to obtain a third vector; select a similarity greater than the second preset similarity in the second vector Degree vector elements (video titles) to obtain a fourth vector.

Wherein, the specific value of the first preset similarity is determined according to the user's interest in commercial shorts, and the specific value of the second preset similarity is determined according to the user's interest in video programs. Thus, the specific value of the first preset similarity and the specific value of the second preset similarity can be set according to the actual situation.

The sub-step S132 is to process the third vector and the fourth vector respectively, and combine the processed third vector and the fourth vector to obtain a mixed recommendation vector.

Please refer to FIG. 5 , which is a schematic flowchart of the sub-steps included in the sub-step S132 in FIG. 4 . The sub-step S132 may include a sub-step S1321 and a sub-step S1322.

The sub-step S1321 is to perform normalization processing on the third vector and the fourth vector respectively.

In the implementation of this embodiment, the similarities corresponding to the vector elements in the third vector and the fourth vector are respectively normalized, so that the advertising clips and video programs can be processed in a unified manner, which is equivalent to Think of the commercial as a video program, too. Among them, normalization is a way to simplify calculations, which refers to transforming a dimensional expression into a dimensionless expression, thus becoming a scalar.

The sub-step S1322 is to combine the normalized third vector and the fourth vector to obtain a mixed vector, and arrange the mixed vectors in descending or ascending order according to the similarity of the vector elements in the mixed vector to obtain a mixed recommendation vector.

In this embodiment, the normalized third vector and fourth vector are combined into one vector to obtain a mixed vector. The mixed vector includes commercial clips and the normalized similarity corresponding to each commercial clip, video programs and the normalized similarity corresponding to each video program. The vector elements in the mixed vector are sorted in descending order or ascending order according to the normalized similarity, so as to obtain a mixed recommendation vector.

The following is an example to introduce how to obtain the mixed recommendation vector.

Please refer to FIG. 6 , which is a schematic diagram of obtaining a mixed recommendation vector provided by a preferred embodiment of the present invention. The first vector and the second vector are obtained from the database 112, wherein the vector elements of the first vector and the second vector are arranged in descending order according to the similarity of the vector elements. The first vector is: [(a1, S1), (a2, S2)..., (am, Sm)], and the second vector is: [(v1, s1), (v2, s2)...,(vn,sn)]. a1, a2, etc. represent short advertisements, S1, S1, etc. represent the similarity between short advertisements and users, m represents the number of short advertisements; v1, v2, etc. represent video programs, s1, s1, etc. represent the similarity between video programs and users, n represents the number of video programs.

Select k advertising clips in the first vector to get the third vector: [(a1,S1),(a2,S2)...,(ak,Sk)]; select h in the second vector video programs to get the fourth vector: [(v1,s1),(v2,s2)...,(vh,sh)]. A normalization process is performed on the third vector and the fourth vector to obtain the normalized third vector and the fourth vector. The third vector after normalization processing is: [(a1,C1),(a2,C2)...,(ak,Ck)]; the fourth vector after normalization processing is: [( v1,c1),(v2,c2)...,(vh,ch)]. Then merge the normalized third vector and the fourth vector, and arrange them in descending order according to the similarity after normalization to get the mixed recommendation vector: [(r1,w1),(r2,w2).... ..,(r(h+k),w(h+k))]. r1, r2, etc. represent commercial clips or video programs, and w1, w2, etc. represent the similarity after normalization processing. Among them, for example, when normalizing s1, s2, ..., sh, first calculate the mean s and standard deviation t, (s1-s)/t is the result of s1 normalization, and so on.

Step S140 , according to the mixed recommendation vector, a mixed recommendation list of advertisement clips and video programs is obtained.

Please refer to FIG. 7 , which is a schematic flowchart of the sub-steps included in step S140 in FIG. 3 . The step S140 may include a sub-step S141 and a sub-step S142.

The sub-step S141 is to search for a commercial clip or video program corresponding to each vector element in the mixed recommendation vector.

In this embodiment, the database 112 also stores correspondences between vector elements and short commercials or video programs. By searching in the database 112, the advertisement clips and video programs included in the mixed recommendation vector can be obtained.

The sub-step S142 is to obtain a mixed recommendation list of commercial shorts and video programs according to the searched commercial shorts and video programs, so that the user terminal 200 plays commercial shorts and video programs based on the mixed recommendation list.

In this embodiment, after the commercial clips and video programs corresponding to the vector elements are obtained, the commercial clips and video programs are sorted in ascending or descending order according to the normalized similarity corresponding to the commercial clips and video programs, so as to obtain A mixed recommendation list of commercial clips and video programs. Therefore, the user terminal 200 can play advertisement clips and video programs according to the mixed recommendation list, thereby improving user experience. In the implementation of this embodiment, the advertisement clips and video programs in the mixed recommendation list are arranged in descending order.

After getting a user's mixed recommendation list of advertising clips and video programs, cyclically calculate the mixed recommendation list of advertising clips and video programs for each user, and then the total mixed recommendation list can be obtained.

Please refer to FIG. 8 . FIG. 8 is another schematic flow chart of a method for recommending a mixture of advertising clips and video programs according to a preferred embodiment of the present invention. The method may also include step S110.

The step S110 is to establish a first vector used to describe the user's degree of interest in the advertisement clip and a second vector used to describe the user's degree of interest in the video program, and store the first vector and the second vector in the database 112 .

Please refer to FIG. 9 , which is a schematic flowchart of some sub-steps included in step S110 in FIG. 8 . The step S110 may include sub-step S116 , sub-step S117 , sub-step S118 and sub-step S119 .

The sub-step S116 is to obtain the advertisement clip vector, the video program vector and the user feature vector, wherein the advertisement clip vector includes one or more tags corresponding to each advertisement clip, and the video program vector includes each video program corresponding The user feature vector includes one or more tags of each video program watched by the user or one or more tags of each commercial clip.

The sub-step S117 is to calculate the similarity between the advertisement clip vector and the user feature vector and the similarity between the video program vector and the user feature vector.

Among them, the similarity between the advertisement video vector and the user feature vector and the similarity between the video program vector and the user feature vector can be calculated by vector inner product, jaccard similarity, Euclidean distance, etc.

The sub-step S118 is to calculate the similarity between each advertising video in the advertising video vector and the user according to the similarity between the advertising video vector and the user feature vector, and according to the advertising video vector and each advertising video The similarity with the user results in a first vector.

Wherein, the first vector includes at least one commercial video and the similarity between each commercial video and the user.

The sub-step S119 is to calculate the similarity between each video program in the video program vector and the user according to the similarity between the video program vector and the user feature vector, and according to the video program vector and each video program The similarity to the user results in a second vector.

Wherein, the second vector includes at least one video program and the similarity between each video program and the user.

Please refer to FIG. 10 , which is a schematic flow chart of another part of sub-steps included in step S110 in FIG. 8 . The step S110 may also include sub-step S112, sub-step S113 and sub-step S114.

The sub-step S112 is to obtain the tag of the commercial clip and the tag of the video program, obtain the vector of the commercial clip according to the tag of the clip, and obtain the vector of the video program according to the tag of the video program.

In this embodiment, the way to convert the commercial clips and video programs into numerical vectors may be to obtain tags of the commercial clips and video programs first, and then convert the tags into numerical vectors. The way to obtain labels can be, but not limited to, manually labeling videos (short commercials, video programs) from various dimensions (such as scenes, shooting techniques, etc.), Automatically extract tags and more. After the tags are obtained, the tags can be converted into numerical vectors through machine learning technology (eg, word2vector) to obtain vectors of commercial clips and video programs, so that the correlation between texts can be better calculated.

In the sub-step S113, the advertisement clips and video programs watched by the user are obtained from the user behavior log.

The sub-step S114 is to extract the tags of the commercial clips or video programs watched by the user, and use the tags of the commercial clips or video programs watched by the user as user tags, so as to obtain the user feature vector.

In this embodiment, the method of obtaining the user feature vector is the same as the method of obtaining the advertisement clip vector and the video program vector, and will not be repeated here.

Please refer to FIG. 11 . FIG. 11 is a schematic block diagram of a device 300 for recommending mixed advertisement clips and video programs according to a preferred embodiment of the present invention. The hybrid recommendation device 300 is applied to the server 100 connected in communication with the user terminal 200 . The server 100 includes a database 112, and the database 112 stores a first vector used to describe the user's degree of interest in the advertisement clip and a second vector used to describe the user's degree of interest in the video program. The hybrid recommendation device 300 includes an acquisition module 320 , a hybrid recommendation vector module 330 and a recommendation list generation module 340 .

The obtaining module 320 is configured to obtain the first vector and the second vector.

In this embodiment, the obtaining module 320 is configured to execute step S120 in FIG. 3 . For a specific description of the obtaining module 320 , refer to the description of step S120 in FIG. 3 .

The mixed recommendation vector module 330 is configured to respectively select vector elements satisfying preset conditions from the first vector and the second vector, and process the selected vector elements to form a mixed recommendation vector.

Please refer to FIG. 12 . FIG. 12 is another schematic block diagram of a device 300 for recommending mixed advertisement clips and video programs according to a preferred embodiment of the present invention. The mixed recommendation vector module 330 includes a selection submodule 331 and a processing submodule 332 .

The selection sub-module 331 is configured to select vector elements satisfying preset conditions from the first vector and the second vector to form a third vector and a fourth vector.

The selection sub-module 331 respectively selects vector elements satisfying preset conditions from the first vector and the second vector to form the third vector and the fourth vector in any one of the following two ways:

In the first vector, a first preset number of vector elements are selected in order of similarity from high to bottom to form a third vector, and in the second vector, a second preset number is selected in order of similarity from high to bottom The vector elements form the fourth vector, wherein, the vector elements in the first vector and the second vector are arranged in descending or ascending order according to the similarity of the vector elements;

Select vector elements whose similarity of vector elements is greater than the first preset similarity in the first vector to form a third vector, and select vector elements whose similarity of vector elements is greater than the second preset similarity in the second vector to form Fourth vector.

The processing sub-module 332 is configured to process the third vector and the fourth vector respectively, and combine the processed third vector and the fourth vector to obtain a mixed recommendation vector.

The processing sub-module 332 processes the third vector and the fourth vector respectively, and merges the processed third vector and the fourth vector to obtain a mixed recommendation vector including:

respectively performing normalization processing on the third vector and the fourth vector;

Combining the normalized third vector and the fourth vector to obtain a mixed vector, and sorting the mixed vectors in descending order or ascending order according to the similarity of vector elements in the mixed vector to obtain a mixed recommendation vector.

In this embodiment, the hybrid recommendation vector module 330 is used to execute step S130 in FIG. 3 . For a specific description of the hybrid recommendation vector module 330 , refer to the description of step S130 in FIG. 3 .

The recommendation list generating module 340 is configured to obtain a mixed recommendation list of commercial shorts and video programs according to the mixed recommendation vector, so that the user terminal 200 can play commercial shorts and video programs based on the mixed recommendation list.

The way in which the recommendation list generation module 340 obtains a mixed recommendation list of commercial clips and video programs according to the mixed recommendation vector includes:

Find the commercial clip or video program corresponding to each vector element in the mixed recommendation vector;

A mixed recommendation list of commercial short films and video programs is obtained according to the searched commercial short films and video programs.

In this embodiment, the recommendation list generation module 340 is configured to execute step S140 in FIG. 3 . For a specific description of the recommendation list generation module 340 , refer to the description of step S140 in FIG. 3 .

Please refer to FIG. 12 again, the hybrid recommendation apparatus 300 may further include a vector creation module 310 . The establishment vector module 310 is used to establish the first vector used to describe the user's degree of interest in the advertising clip, the second vector used to describe the degree of interest in the video program, and save the first vector and the second vector in the database 112 in.

The establishment vector module 310 establishes a first vector used to describe the user's degree of interest in the advertising clip, a second vector used to describe the degree of interest in the video program, and saves the first vector and the second vector in the database 112 Ways include:

Obtaining the advertisement clip vector, the video program vector and the user feature vector, wherein the advertisement clip vector includes one or more labels corresponding to each advertisement clip, and the video program vector includes one or more tags corresponding to each video program , the user feature vector includes one or more tags of each video program watched by the user or one or more tags of each commercial clip;

Calculating the similarity between the advertisement clip vector and the user feature vector and the similarity between the video program vector and the user feature vector;

Calculate the similarity between each advertisement clip in the advertisement clip vector and the user according to the similarity between the advertisement clip vector and the user feature vector, and calculate the similarity between the advertisement clip vector and each advertisement clip and the user get the first vector;

Calculate the similarity between each video program in the video program vector and the user according to the similarity between the video program vector and the user feature vector, and calculate the similarity between the video program vector and each video program and the user according to the similarity between the video program vector and the user feature vector to get the second vector.

The establishment vector module 310 establishes a first vector used to describe the user's degree of interest in the advertising clip, a second vector used to describe the degree of interest in the video program, and saves the first vector and the second vector in the database 112 Ways also include:

Obtain the label of the commercial video and the label of the video program, obtain the commercial video vector according to the label of the commercial video, and obtain the video program vector according to the label of the video program;

Obtain the short advertisements and video programs that the user has watched from the user behavior log;

The tags of the commercial clips or video programs watched by the user are extracted, and the tags of the commercial clips or video programs watched by the user are used as user tags, so as to obtain the user feature vector.

In this embodiment, the vector creation module 310 is configured to execute step S110 in FIG. 8 . For a specific description of the vector creation module 310 , reference may be made to the description of step S110 in FIG. 8 .

To sum up, the present invention provides a method and device for recommending mixed advertisement clips and video programs, which is applied to a server connected to a user terminal through communication. The server includes a database, and the database stores a first vector used to describe the user's degree of interest in the advertisement clip and a second vector used to describe the degree of user's interest in the video program. After the first vector and the second vector are acquired, vector elements in the first vector and the second vector are selected according to preset conditions, and the selected vector elements are processed to obtain a mixed recommendation vector. Thus, a mixed recommendation list of commercial shorts and video programs is obtained by mixing recommendation vectors, and the user terminal plays commercial shorts and video programs according to the mixed recommendation list. Therefore, the commercial short film and the video program are treated equally, and a unified method is used to process the mixed recommendation of the commercial short film and the video program, instead of inserting the commercial short film into the video program. In this manner, users will not be disturbed by short advertisements when watching video programs, and a good balance is achieved between the income of short advertisements and user experience.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. A method for recommending mixed advertisement clips and video programs, which is applied to a server connected in communication with a user terminal, characterized in that the server includes a database, and the database stores first information for describing the degree of interest of the user in the advertisement clips. vector, a second vector used to describe the degree of interest in the video program, the method comprising: obtaining the first vector and the second vector; Select vector elements satisfying preset conditions from the first vector and the second vector respectively to form a third vector and a fourth vector; perform normalization processing on the third vector and the fourth vector respectively; normalize Merging the processed third vector and the fourth vector to obtain a mixed vector, and sorting the mixed vectors in descending order or ascending order according to the similarity of vector elements in the mixed vector to obtain a mixed recommendation vector; A mixed recommendation list of commercial short films and video programs is obtained according to the mixed recommendation vectors, so that the user terminal plays commercial short films and video programs based on the mixed recommendation list. 2. The method according to claim 1, wherein the method of selecting vector elements satisfying preset conditions from the first vector and the second vector to form the third vector and the fourth vector is as follows: any one of: In the first vector, a first preset number of vector elements are selected in order of similarity from high to bottom to form a third vector, and in the second vector, a second preset number is selected in order of similarity from high to bottom The vector elements form the fourth vector, wherein, the vector elements in the first vector and the second vector are arranged in descending or ascending order according to the similarity of the vector elements; Select vector elements whose similarity of vector elements is greater than the first preset similarity in the first vector to form a third vector, and select vector elements whose similarity of vector elements is greater than the second preset similarity in the second vector to form Fourth vector. 3. The method according to claim 1, further comprising: establishing a first vector used to describe the user's degree of interest in the commercial clip, a second vector used to describe the degree of interest in the video program, and saving the first vector and the second vector in a database, The step of establishing the first vector used to describe the user's degree of interest in the advertising clip, the second vector used to describe the degree of interest in the video program, and storing the first vector and the second vector in the database includes: Obtaining the advertisement clip vector, the video program vector and the user feature vector, wherein the advertisement clip vector includes one or more labels corresponding to each advertisement clip, and the video program vector includes one or more tags corresponding to each video program , the user feature vector includes one or more tags of each video program watched by the user or one or more tags of each commercial clip; Calculating the similarity between the advertisement clip vector and the user feature vector and the similarity between the video program vector and the user feature vector; Calculate the similarity between each advertisement clip in the advertisement clip vector and the user according to the similarity between the advertisement clip vector and the user feature vector, and calculate the similarity between the advertisement clip vector and each advertisement clip and the user get the first vector; Calculate the similarity between each video program in the video program vector and the user according to the similarity between the video program vector and the user feature vector, and calculate the similarity between the video program vector and each video program and the user according to the similarity between the video program vector and the user feature vector to get the second vector. 4. The method according to claim 3, characterized in that, said establishment is used to describe the first vector of the user's degree of interest in the advertising clip, the second vector used to describe the degree of interest in the video program, and the first vector The step of saving the first vector and the second vector in the database also includes: Obtain the label of the commercial video and the label of the video program, obtain the commercial video vector according to the label of the commercial video, and obtain the video program vector according to the label of the video program; Obtain the short advertisements and video programs that the user has watched from the user behavior log; The tags of the commercial clips or video programs watched by the user are extracted, and the tags of the commercial clips or video programs watched by the user are used as user tags, so as to obtain the user feature vector. 5. The method according to claim 1, wherein the step of obtaining a mixed recommendation list of advertising clips and video programs according to the mixed recommendation vector comprises: Find the commercial clip or video program corresponding to each vector element in the mixed recommendation vector; A mixed recommendation list of commercial short films and video programs is obtained according to the searched commercial short films and video programs. 6. A device for recommending mixed advertising clips and video programs, which is applied to a server connected to a user terminal in communication, wherein the server includes a database, and the database stores first information describing the degree of interest of the user in the advertising clips. vector, a second vector used to describe the degree of interest in the video program, the device includes: An acquisition module, configured to acquire the first vector and the second vector; The mixed recommendation vector module is used to select vector elements that meet the preset conditions from the first vector and the second vector to form a third vector and a fourth vector; respectively normalize the third vector and the fourth vector processing; merging the normalized third vector and the fourth vector to obtain a mixed vector, and sorting the mixed vector in descending or ascending order according to the similarity of the vector elements in the mixed vector to obtain a mixed recommendation vector; The recommendation list generating module is configured to obtain a mixed recommendation list of commercial short films and video programs according to the mixed recommendation vector, so that the user terminal can play commercial short films and video programs based on the mixed recommendation list. 7. The device according to claim 6, wherein the hybrid recommendation vector module selects vector elements satisfying preset conditions from the first vector and the second vector to form the third vector and the fourth vector One of the following two methods: In the first vector, a first preset number of vector elements are selected in order of similarity from high to bottom to form a third vector, and in the second vector, a second preset number is selected in order of similarity from high to bottom The vector elements form the fourth vector, wherein, the vector elements in the first vector and the second vector are arranged in descending or ascending order according to the similarity of the vector elements; Select vector elements whose similarity of vector elements is greater than the first preset similarity in the first vector to form a third vector, and select vector elements whose similarity of vector elements is greater than the second preset similarity in the second vector to form Fourth vector. 8. The device according to claim 6, characterized in that the device further comprises: a vector building module, the vector building module is used to create a first vector for describing the user's degree of interest in the advertising clip, for describing a second vector of the degree of interest in the video program, and storing the first vector and the second vector in a database, The vector creation module establishes a first vector used to describe the user's degree of interest in the advertisement clip, a second vector used to describe the degree of interest in the video program, and saves the first vector and the second vector in the database include: Obtaining the advertisement clip vector, the video program vector and the user feature vector, wherein the advertisement clip vector includes one or more labels corresponding to each advertisement clip, and the video program vector includes one or more tags corresponding to each video program , the user feature vector includes one or more tags of each video program watched by the user or one or more tags of each commercial clip; Calculating the similarity between the advertisement clip vector and the user feature vector and the similarity between the video program vector and the user feature vector; Calculate the similarity between each advertisement clip in the advertisement clip vector and the user according to the similarity between the advertisement clip vector and the user feature vector, and calculate the similarity between the advertisement clip vector and each advertisement clip and the user get the first vector; Calculate the similarity between each video program in the video program vector and the user according to the similarity between the video program vector and the user feature vector, and calculate the similarity between the video program vector and each video program and the user according to the similarity between the video program vector and the user feature vector to get the second vector. 9. The device according to claim 8, wherein the vector building module creates a first vector for describing the user's degree of interest in the commercial clip, a second vector for describing the degree of interest in the video program, and The manner in which the first vector and the second vector are stored in the database also includes: Obtain the label of the commercial video and the label of the video program, obtain the commercial video vector according to the label of the commercial video, and obtain the video program vector according to the label of the video program; Obtain the short advertisements and video programs that the user has watched from the user behavior log; The tags of the commercial clips or video programs watched by the user are extracted, and the tags of the commercial clips or video programs watched by the user are used as user tags, so as to obtain the user feature vector. 10. The device according to claim 6, wherein the method for the recommendation list generating module to obtain a mixed recommendation list of advertising clips and video programs according to the mixed recommendation vector comprises: Find the commercial clip or video program corresponding to each vector element in the mixed recommendation vector; A mixed recommendation list of commercial short films and video programs is obtained according to the searched commercial short films and video programs.