Media content pushing method, device and equipment, and storage medium

Abstract

Embodiments of the present application provide a media content pushing method and device, equipment and storage medium, and relate to the technical field of computer and Internet. The method comprises: obtaining a viewing record of a target object on first media content, the viewing record being used to reflect a viewing preference of the target object on second media content; wherein the first media content and the second media content belong to two different fields; determining a recommendation score of each of n second media contents in each recommendation strategy according to the viewing record, n being a positive integer; determining a pushing order of the n second media contents based on the recommendation score of each of the n second media contents in each recommendation strategy; and pushing the second media content to the target object according to the pushing order. The technical solution provided by the embodiments of the present application can improve the accuracy of media content pushing.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of computer and Internet, and particularly relate to a media content pushing method and device, equipment and storage medium. BACKGROUND

[0002] With the increasing time of people using the Internet, the pushing mode of pushing personalized media content to different users based on user interests is also developing.

[0003] In the related technology, when a user needs to obtain news content, a pushing algorithm program pushes news content to the user according to the browsing record of the user on the news content. In the above related technology, when the browsing record of the user on the news content cannot be obtained in a sufficient amount, the news content interested by the user cannot be accurately pushed to the user. SUMMARY

[0004] Embodiments of the present application provide a media content pushing method, device, equipment and storage medium, which can improve the accuracy of media content pushing. The technical solution is as follows:

[0005] According to an aspect of an embodiment of the present application, a media content pushing method is provided, the method comprising:

[0006] obtaining a viewing record of a target object on first media content, the viewing record being used to reflect the viewing preference of the target object on second media content; wherein the first media content and the second media content belong to two different fields;

[0007] determining the recommendation score of n items of second media content in each recommendation strategy according to the viewing record, n being a positive integer;

[0008] determining the pushing order of the n items of second media content based on the recommendation score of the n items of second media content in each recommendation strategy;

[0009] pushing the second media content to the target object according to the pushing order.

[0010] According to an aspect of an embodiment of the present application, a media content pushing device is provided, the device comprising:

[0011] a record obtaining module configured to obtain a viewing record of a target object on first media content, the viewing record being used to reflect the viewing preference of the target object on second media content; wherein the first media content and the second media content belong to two different fields;

[0012] A score determination module is configured to determine, according to the viewing record, a recommendation score of each of the n second media contents in each of the recommendation strategies, where n is a positive integer;

[0013] A sorting determination module is configured to determine, based on the recommendation score of each of the n second media contents in each of the recommendation strategies, a push sorting of the n second media contents.

[0014] A content push module is configured to push the second media contents to the target object according to the push sorting.

[0015] According to an aspect of an embodiment of the present application, a computer device is provided, which includes a processor and a memory, and the memory stores a computer program, which is loaded and executed by the processor to implement the above media content push method.

[0016] According to an aspect of an embodiment of the present application, a computer readable storage medium is provided, which stores a computer program, and the computer program is loaded and executed by a processor to implement the above media content push method.

[0017] According to an aspect of an embodiment of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the above media content push method.

[0018] The technical solution provided by the embodiments of the present application can have the following beneficial effects:

[0019] By determining the push sorting of the second media contents to the target object according to the viewing record of the target object on the first media content, the viewing record is used to reflect the viewing preference of the target object on the second media content, so as to transfer the behavior portrait of the target object in one field to another field, to realize the cross-field push of the second media content that the target object is more interested in to the target object, and to improve the accuracy of media content push in cold start.

[0020] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description only represent some of the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

[0022] Figure 1 is a schematic diagram of a media content pushing system provided by an embodiment of the present application;

[0023] Figure 2 is a flowchart of a media content pushing method provided by an embodiment of the present application;

[0024] Figure 3 is a schematic diagram of a user interface provided by an embodiment of the present application;

[0025] Figure 4 is a flowchart of a media content pushing method provided by another embodiment of the present application;

[0026] Figure 5 is a flowchart of a media content pushing method provided by another embodiment of the present application;

[0027] Figure 6 is a block diagram of a media content pushing apparatus provided by an embodiment of the present application;

[0028] Figure 7 is a block diagram of a media content pushing apparatus provided by another embodiment of the present application;

[0029] Figure 8 is a block diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0030] The exemplary embodiments will be described in detail herein with reference to the drawings. Unless otherwise specified, the same numbers in different drawings indicate the same or similar elements. The following exemplary embodiments described in the following description are not meant to be exhaustive or to be limited to the precise form of the applications disclosed. They are presented for the purpose of best explaining and providing the full scope of the applications as set forth in the appended claims.

[0031] Artificial Intelligence (AI) is the theory, method, technology and application system of using digital computer or machine controlled by digital computer to simulate, extend and expand human intelligence, perceive environment, acquire knowledge and use knowledge to obtain the best results. Artificial intelligence technology involves a wide range of fields, both hardware and software level technology. Artificial intelligence basic technology generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0032] Machine Learning (ML) is a multi-disciplinary subject, involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is a subject that studies how computers simulate or implement human learning behavior to acquire new knowledge or skills, and reorganize existing knowledge structure to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental approach to making computers intelligent, and its applications are widespread in various fields of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning technologies.

[0033] The scheme provided by the embodiments of the present application relates to machine learning technology of artificial intelligence, for example, using machine learning technology to push second media content to a user based on the viewing record of first media content.

[0034] Please refer to Figure 1 , which shows a schematic diagram of a media content pushing system provided by an embodiment of the present application. As Figure 1 shown, the system 10 includes a terminal 11 and a server 12. Among them, the terminal 11 is a terminal used by a target object, and the terminal 11 and the server 12 establish a network connection, the terminal 11 is used to send the viewing record of the target object to the first media content to the server 12; the server 12 is used to determine the push order of n second media content based on the viewing record of the target object to the first media content, and push the sorted second media content to the terminal 11; the terminal 11 is also used to display the second media content according to the push order.

[0035] In some embodiments, the system 10 further comprises at least one terminal 13, the terminal 13 being a terminal used by a first object other than the target object, and the terminal 13 is configured to send a viewing record of the first object for the first media content and the second media content to the server 12, and the server 12 is capable of determining a push order of the n items of second media content based on the viewing record of each first object for the first media content and the second media content, and the viewing record of the target object for the first media content, and pushing the sorted second media content to the terminal 11.

[0036] The terminal 11 and the server 12 are electronic devices with data computing, processing and storage capabilities. Among them, the terminal 11 can be a smart phone, a PC (Personal Computer), a tablet computer, a wearable device, a smart television, a smart robot, etc.; the server 12 can be a stand-alone physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The execution subject of each step of the method provided by the embodiments of the present application can be the terminal 11 or the server 12; in other embodiments, each step of the method provided by the embodiments of the present application can also be executed by the terminal 11 and the server 12 in interaction.

[0037] In the following embodiments, in order to simplify the content, only the execution subject of each step is a computer device to introduce the embodiments of the present application.

[0038] The cloud server is a server deployment method based on cloud technology. Cloud technology refers to a series of resources such as hardware, software, network, etc. unified in a wide area network or local area network to realize data computing, storage, processing and sharing.

[0039] Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on cloud computing business model application, which can form a resource pool, and can be used on demand, flexibly and conveniently. Cloud computing technology will become an important support. Cloud servers are used to provide background services for technical network systems, requiring a large amount of computing and storage resources, such as video websites, image websites and more portals. With the high development and application of the Internet industry, every item may have its own identification mark in the future, and it needs to be transmitted to the background system for logical processing. Different levels of data will be processed separately, and data of various industries will need strong system support, which can only be realized through cloud computing. Cloud-based technologies include cloud computing, cloud storage, databases, big data, etc.

[0040] With the development and progress of the industry, cloud technology will play an increasingly important role in various industries.

[0041] The technical solutions of the present application will be described below through several embodiments.

[0042] Please refer to Figure 2 which shows a flowchart of a media content pushing method provided by an embodiment of the present application. In the embodiment, the method is mainly exemplified by being applied to the computer device described above. The method can include the following steps (201-204):

[0043] Step 201, obtaining a viewing record of a target object on a first media content, the viewing record being used to reflect a viewing preference of the target object on a second media content.

[0044] The target object includes a target account of an application program or a portal website, and the viewing record is a record generated by a user of the target account when viewing the first media content.

[0045] In some embodiments, the first media content and the second media content belong to two different fields. Optionally, the field to which the first media content belongs is any one of the following: film and television, news, short video, text and image, live broadcast, and question and answer; and the field to which the second media content belongs is any one of the following: film and television, news, short video, text and image, live broadcast, and question and answer. As shown in Figure 3 When the second media content is news content, the display form of the news content can be a video form in the user interface 30. Of course, the news content can also be displayed in the form of text, picture, audio, etc. The first media content and the second media content can be contents in the same application program or the same portal website; or the first media content and the second media content can be contents in different application programs or different portal websites. Optionally, the first media content is film and television content, and the viewing record is a viewing record of the target object.

[0046] In some embodiments, the first media content and the second media content have relevance. For example, the classification of the first media content is the same as or similar to the classification of the second media content. In an example, the classification of the first media content includes: life, emotion, idol, humor, education, city, sports, etc., and the classification of the second media content includes: society, emotion, entertainment, humor, education, finance, sports, etc. The viewing record of the target object can represent its preference for part of the first media content, which has a high probability of coincidence with the preference of the target object for the second media content. For example, if there are more idol-type first media contents in the viewing record, the target object is likely to be more interested in the entertainment-type second media content. Therefore, the viewing record can reflect the viewing preference of the target object for the second media content.

[0047] Step 202, determining, according to the viewing record, a recommendation score of n items of second media content in each recommendation strategy respectively, n being a positive integer.

[0048] In some embodiments, a plurality of recommendation strategies are obtained, and the plurality of recommendation strategies are used to represent degrees of association between each item of second media content and the viewing record from different dimensions. Thus, based on the viewing record, a recommendation score of each item of second media content in each recommendation strategy can be determined, and the recommendation score is used to indicate a degree of recommendation of the item of second media content in the corresponding recommendation strategy.

[0049] In step 203, a push order of the n items of second media content is determined based on the recommendation scores of the n items of second media content in each recommendation strategy.

[0050] After the recommendation scores of the n items of second media content in each recommendation strategy are obtained, a push order of the n items of second media content can be determined according to the final push strategy. Alternatively, the n items of second media content are contents that have not been pushed to the target object or have not been viewed by the target object. In some embodiments, the push order is updated once every x times of updating of the viewing record, and x is a positive integer.

[0051] In some embodiments, step 203 further includes the following sub-steps:

[0052] 1. For the n items of second media content, a maximum value of the recommendation scores of the i-th item of second media content in each push strategy is determined as a final recommendation score of the i-th item of second media content, and i is a positive integer.

[0053] 2. The push order is determined according to the respective latest recommendation scores of the n items of second media content.

[0054] In this implementation, for the i-th item of second media content in the n items of second media content, a maximum value corresponding to the i-th item of second media content in each recommendation strategy is determined as a final recommendation score of the i-th item of second media content. Then, the n items of second media content are sorted in descending order of the respective final recommendation scores, and are pushed in descending order of the final recommendation scores, for example, the media content with the highest final recommendation score is preferentially pushed, then the media content with the highest final recommendation score in the remaining second media content is pushed, and so on. For another example, in the n items of second media content sorted in descending order of the final recommendation scores, the first k items of second media content are preferentially pushed to the target object, then the first k items of second media content in the remaining second media content are pushed, and so on, and k is a positive integer greater than 1. Alternatively, the second media content with a higher final recommendation score in the k items of second media content pushed each time is preferentially displayed.

[0055] In some embodiments, step 203 further includes the following sub-steps:

[0056] 1. A priority of each recommendation strategy is obtained.

[0057] 2. determining the push order according to the recommendation scores of the n second media contents in each push strategy respectively and the priority of each push strategy.

[0058] The priority of each push strategy is determined in advance, and the n second media contents are sorted according to the recommendation scores obtained by the push strategies with higher priority first. That is, the n second media contents are sorted according to the recommendation scores of the push strategy with the highest priority first, and for the second media contents with equal recommendation scores of the push strategy with the highest priority, the n second media contents are sorted according to the recommendation scores of the push strategy with the second highest priority, and so on.

[0059] In an example, the order of the push strategies according to the priority is: push strategy 1> push strategy 2> push strategy 3; n is 3, and the recommendation scores of the 3 second media contents corresponding to each push strategy are shown in Table 1.

[0060] Table 1

[0061]

[0062] Since the priority of the push strategy 1 is the highest, the 3 second media contents are sorted according to the recommendation scores of the push strategy 1 from high to low first: second media content 3> second media content 1= second media content 2; since the recommendation scores of second media content 1 and second media content 2 are equal according to the recommendation scores of the push strategy 1, the second media content 1 and the second media content 2 are sorted according to the recommendation scores of the push strategy 2 from high to low: second media content 2> second media content 1. Then the push order of the 3 second media contents is: second media content 3> second media content 2> second media content 1.

[0063] Alternatively, the priority of each push strategy can be set by a relevant technical staff according to actual conditions, which is not limited in the embodiments of the present application. In this implementation, the n second media contents are sorted according to the recommendation scores of the n second media contents in each push strategy respectively and the priority of each push strategy, so as to ensure that the second media content with a higher recommendation score in the push strategy which is most concerned by the relevant staff can be pushed first.

[0064] In some embodiments, step 203 further includes the following sub-steps:

[0065] 1. obtaining the recommendation weight corresponding to each push strategy respectively;

[0066] 2. determining the final recommendation score corresponding to the n second media contents respectively according to the recommendation scores of the n second media contents in each push strategy respectively and the recommendation weight corresponding to each push strategy respectively.

[0067] 3. Determine the push order according to the final recommendation scores of the n items of second media content respectively corresponding to the second media content.

[0068] In this implementation, each recommendation strategy corresponds to a recommendation weight. Optionally, the final recommendation score of the second media content is obtained by weighted summation of the recommendation scores obtained by each recommendation strategy, and the n items of second media content are sorted in descending order of the final recommendation score, i.e., the push order is obtained.

[0069] In some embodiments, the final recommendation score is calculated according to the following Formula One:

[0070] Formula One:

[0071]

[0072] wherein R is the final recommendation score, A n is the recommendation weight of the nth recommendation strategy, and R n is the recommendation score of the nth recommendation strategy.

[0073] In one example, three recommendation strategies and the recommendation scores of the three items of second media content respectively corresponding to each recommendation strategy are shown in Table 1 above, and the recommendation weights of the three recommendation strategies (recommendation strategy 1, recommendation strategy 2, and recommendation strategy 3) are 0.5, 0.3, and 0.2 respectively. Then the final recommendation score of the second media content 1 is 0.5x5+0.3x4+0.2x6=4.9; the final recommendation score of the second media content 2 is 0.5x5+0.3x9+0.2x3=5.8; and the final recommendation score of the second media content 3 is 0.5x7+0.3x6+0.2x2=5.7, and the push order is second media content 2>second media content 3>second media content 1.

[0074] Step 204, push the second media content to the target object according to the push order.

[0075] After determining the push order, the second media content is pushed to the target object according to the push order, and the second media content at the front of the push order is preferentially displayed. Optionally, the push content at the front of the push order is displayed in a more conspicuous manner, such as being displayed at the top of the display interface, being enlarged in display area, being extended in display time, etc.

[0076] To sum up, in the technical solution provided in the embodiment of the present application, the push order of the second media content to the target object is determined based on the target object's viewing record of the first media content. The viewing record is used to reflect the target object's viewing preference for the second media content, thereby migrating the target object's behavioral portrait in one field to another field, and realizing the cross-field push of the second media content that the target object is more interested in, thereby improving the accuracy of media content push during cold start.

[0077] For the recommended strategies mentioned above, the following introduces several recommended strategies as examples.

[0078] Recommended strategy 1:

[0079] In some embodiments, the above step 202 further includes the following sub-steps (2021-2023):

[0080] Step 2021: Obtain association weights between each category of the second media content and each category of the first media content.

[0081] The association weight is used to indicate the degree of association between the classification of the second media content and the classification of the first media content. In some embodiments, the classifications of the first media content and the second media content are the same or similar. Optionally, the association weight can represent the probability that the target object is interested in both a classification of the first media content and a classification of the second media content.

[0082] In some embodiments, the association weight is set by relevant technical personnel based on actual conditions, and the embodiments of this application do not limit this.

[0083] Step 2022: Determine, based on the viewing records and the associated weights, first recommendation scores for each category of the second media content corresponding to the target object.

[0084] The first recommendation score indicates the target subject's interest in the category of the second media content. The higher the first recommendation score, the higher the target subject's interest in the category of the second media content. For example, if the target subject watches more campus and variety shows in their viewing history, it can be assumed that the target subject is more interested in entertainment news content, and thus the first recommendation score corresponding to entertainment news content is higher.

[0085] In some embodiments, the first media content is film and television content, the second media content is news content, the olive record is the target object's viewing record of the film content, and the calculation formula for each category of the news content corresponding to the target object is as follows: Formula 2:

[0086] Formula 2:

[0087]

[0088] Among them, R 分类 Indicates the first recommendation score of the corresponding category of film and television content, N represents the total number of films watched in the viewing record, n 分类 represents the number of news content corresponding to the film category, W 分类 Represents the association weight of the movie classification.

[0089] In one example, the association weights between various categories of film and television content and various categories of news content are shown in Table 2 below:

[0090] Table 2

[0091]

[0092] For example, the target audience has a total of 10 movies in their viewing history, including 4 variety shows, 2 campus movies, and 4 science fiction movies. Based on Table 2 above, the first recommendation value of each category of news content is calculated as follows:

[0093] The first recommendation score of entertainment news content REntertainment = (4x9+2x7) / 10 = 5;

[0094] The first recommendation score of science and technology news content Rtechnology = (4x6) / 10 = 2.4;

[0095] The first recommendation score of life news content Rlife=(4x5) / 10=2;

[0096] The first recommendation score of the news content of the remaining categories is 0. Based on the calculation and comparison, it can be seen that in this example, the target object tends to browse entertainment news content, followed by technology news content, and then life news content.

[0097] Step 2023: Determine first recommendation scores of n items of second media content based on the first recommendation scores of the target objects corresponding to the categories of the second media content.

[0098] The first recommendation score for the i-th second media content item among the n second media content items is the first recommendation score for the target object in the category to which the i-th second media content item belongs, where i is a positive integer. For example, in the above example, the first recommendation score REntertainment = 5 for entertainment news content. If the i-th second media content item is entertainment news content, then the first recommendation score for the i-th second media content item is 5.

[0099] Recommended strategy 2:

[0100] In some embodiments, the above step 202 further includes the following sub-steps (2024-2025):

[0101] Step 2024, obtaining, from the viewing record, a viewing record of the target object for the popular first media content.

[0102] The popular first media content refers to the first media content meeting the first popularity condition. The viewing record of the target object for the popular first media content is obtained, that is, the popular first media content and related information viewed in the viewing record are obtained.

[0103] Optionally, the first popularity condition is that the viewing quantity of the first media content within a first time length before viewing is greater than or equal to a first viewing quantity threshold, where the first time length can be 10 minutes, 30 minutes, 1 hour, 2 hours, 3 hours, 6 hours, 1 day, 3 days, 1 week, 2 weeks, etc. The specific time length of the first time length can be set by the relevant technical personnel according to the actual situation, and the present embodiment does not limit this. The specific value of the first viewing quantity threshold can also be set by the relevant technical personnel according to the actual situation, and the present embodiment does not limit this.

[0104] Optionally, the first popularity condition can also be that the Internet discussion degree within a second time length before viewing is greater than or equal to a first discussion degree threshold, where the second time length can be 10 minutes, 30 minutes, 1 hour, 2 hours, 3 hours, 6 hours, 1 day, 3 days, 1 week, 2 weeks, etc. The specific time length of the second time length can be set by the relevant technical personnel according to the actual situation, and the present embodiment does not limit this. The specific value of the second discussion degree threshold can also be set by the relevant technical personnel according to the actual situation, and the present embodiment does not limit this.

[0105] Optionally, the first popularity condition is to view the first media content through a popular recommendation block of an application program or a portal website.

[0106] Step 2025, determining a target value according to the viewing record of the target object for the popular first media content.

[0107] The target value refers to a recommendation score of the popular second media content corresponding to the target object, and the popular second media content refers to the second media content meeting the second popularity condition. For the i th second media content in n second media contents, if the i th second media content belongs to the popular second media content, the second recommendation score of the i th second media content is determined as the target value, and i is a positive integer.

[0108] The introduction to the second popularity condition can refer to the content of step 2024 described above, which will not be repeated here.

[0109] In some embodiments, the first media content is movie content, the popular first media content includes popular movies and popular episodes; the second media content is news content, the popular second media content includes popular news, and the sub-step 2025 further includes the following sub-steps:

[0110] 1. obtaining a first score based on a record of watching a popular movie in the viewing record of the popular first media content;

[0111] 2. determining a second score based on an average duration from starting to play each episode of a popular episode to watching the end in the viewing record of the popular first media content, and a number of popular episodes watched in a first preset time period;

[0112] 3. determining a target value according to the first score and the second score.

[0113] The first score can be determined by a number of popular movies watched in the viewing record. The second score can be obtained by a speed of watching updated popular first media content by the target object. The target value obtained by combining the first score and the second score can indicate a degree of pursuing popular first media content by the target object.

[0114] In one example, the first score can be calculated according to the following Formula Three:

[0115] Formula Three:

[0116]

[0117] Wherein, n is a number of popular movies watched in the viewing record, and M is the first score.

[0118] In one example, the second score can be calculated according to the following Formula Four:

[0119] Formula Four:

[0120]

[0121] Wherein, t is an average number of days from each episode update to watching for a popular episode in the viewing record, X is a preset average day threshold, and N is the second score.

[0122] In one example, the target value can be calculated according to the following Formula Five:

[0123] Formula Five:

[0124] R=M×N

[0125] Wherein, M is the first score, N is the second score, and R is the target value.

[0126] Recommendation strategy three:

[0127] In some embodiments, the step 202 further comprises the following sub-steps (2026-2028):

[0128] In the step 2026, the viewing similarity weight of each first object is determined respectively by viewing records.

[0129] The viewing similarity weight of the mth first object is used to represent the similarity between the viewing record of the mth first object and the viewing record, and m is a positive integer. That is, the viewing similarity weight is used to indicate the importance of the viewing record of each first object in the recommendation strategy from the perspective of the similarity of the viewing record.

[0130] In the step 2027, the viewing record of each first object on the second media content is obtained.

[0131] The viewing record of each first object on the second media content is obtained, and the viewing record of each first object on the second media content is obtained, so that the order of pushing each item of second media content to the target object can be obtained.

[0132] It should be noted that the execution order of the steps 2026 and 2027 can be exchanged.

[0133] In the step 2028, the third recommendation score of each item of second media content is determined according to the viewing similarity weight and the viewing record of each first object on the second media content.

[0134] After determining the viewing similarity weight of each first object and the viewing record of each first object on the second media content, the third recommendation score of each item of second media content is calculated, wherein the higher the viewing similarity weight of the first object corresponding to the second media content, and the more the first object viewing the second media content, the higher the third recommendation score.

[0135] In some embodiments, the sub-step 2028 further comprises the following sub-steps:

[0136] 1. The viewing similarity between the viewing record of each first object and the viewing record is calculated respectively;

[0137] 2. The viewing similarity weight corresponding to each first object is determined based on the viewing similarity corresponding to each first object.

[0138] In some embodiments, the similarity between the viewing record of each first object and the viewing record of the target object is determined first to obtain the viewing similarity, and then the viewing similarity weight corresponding to each first object is determined according to the viewing similarity corresponding to each first object.

[0139] In some embodiments, the calculation of the viewing similarity can refer to the following Formula Six:

[0140] Formula Six:

[0141]

[0142] wherein S is the viewing similarity, K is the number of the same first media content in the viewing records of the first object and the target object, M is a constant, used to represent the M items of the first media content recently viewed by the target object, and L is the number of the first media content recently viewed by the first object.

[0143] In some embodiments, the calculation of the viewing similarity weight can refer to the following Formula Seven:

[0144] Formula Seven:

[0145]

[0146] wherein P m is the viewing similarity weight corresponding to the mth first object, S m is the viewing similarity corresponding to the mth first object, and S n is the viewing similarity corresponding to the first object.

[0147] In one example, the first media content is a film and television content, the second media content is a news content, the viewing record is a film viewing record, M = 10, the film viewing record of the target object is {Film A, Film B, Film C, Film F, Film G, Film H, Film I, Film J, Film K, Film L}, the film viewing record of the first object A is {Film A, Film B, Film D, Film E}, and the viewing similarity S between the first object A and the target object is 2 / ((10+4) / 2) = 2 / 7.

[0148] In another example, the viewing similarities corresponding to the first object A, the first object B and the first object C are 2 / 7, 3 / 10 and 1 / 3 respectively; the viewing similarity weights corresponding to the first object A, the first object B and the first object C are 60 / 193, 63 / 193 and 70 / 193 respectively, which are obtained by using the above Formula Six. Assuming that the browsing record of the news content of the first object A is {News A, News B, News C}, the browsing record of the news content of the first object B is {News A, News D, News E}, and the browsing record of the news content of the first object C is {News B, News C, News E}, the third recommendation scores corresponding to each news content are:

[0149] The third recommendation score of News A = 10*(60 / 193+63 / 193) = 1230 / 193

[0150] The third recommendation score of News B = 10 * (60 / 193 + 70 / 193) = 1300 / 193

[0151] The third recommendation score of News C = 10 * (60 / 193 + 70 / 193) = 1300 / 193

[0152] The third recommendation score of News D = 10 * 63 / 193 = 630 / 193

[0153] The third recommendation score of News E = 10 * (63 / 193 + 70 / 193) = 1330 / 193

[0154] Recommendation strategy four:

[0155] In some embodiments, the step 202 further includes: for the i-th second media content in the n second media contents, if the i-th second media content is real-time second media content, determining the fourth recommendation score of the i-th second media content as a first value; wherein the real-time second media content is second media content generated within a second preset time period.

[0156] In this implementation, the real-time second media content is given a fourth recommendation score, which is used to improve the push ranking of the real-time second media content, such as preferentially pushing real-time news, so that the target object can obtain the latest news content in time. The specific value of the first value is set by the relevant technical personnel according to the actual situation, and the embodiments of the present application are not limited thereto. Optionally, only the real-time news of great importance is given a fourth recommendation score, which is determined by the relevant technical personnel.

[0157] Recommendation strategy five:

[0158] In some embodiments, the step 202 further includes: for the i-th second media content in the n second media contents, if the i-th second media content is related second media content, the fourth recommendation score of the i-th second media content is a second value, and the related second media content is second media content related to the first media content contained in the viewing record; wherein i is a positive integer.

[0159] In this implementation, the target object is pushed the second media content related to the first media content contained in the viewing record, which can enable the target object to obtain relevant information about the first media content in time and improve the accuracy of media content pushing. Whether the i-th second media content is related second media content can be determined by the relevant technical personnel or a trained machine learning model.

[0160] The above-mentioned several strategies are only exemplary, and in actual application process, the relevant technical personnel can also increase or decrease according to the actual situation.

[0161] In one example, the first media content is movie content, and the second media content is news content. As shown, the push order 54 of the news content in the news pool 53 is determined through the movie viewing record 52 of the target object 51, and the news content is pushed to the target object according to the push order 54, thereby realizing cross-domain media content pushing. In another example, the first media content is news content, and the second media content is movie content. The first media content and the second media content can also be content in other domains, and the embodiments of the present application do not limit the same. Figure 5

[0162] The following is an apparatus embodiment of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the apparatus embodiments of the present application, please refer to the method embodiments of the present application.

[0163] Please refer to Figure 6 which shows a block diagram of a media content pushing apparatus provided by one embodiment of the present application. The apparatus has the function of realizing the method examples of the above-mentioned media content pushing. The apparatus 600 can include a record acquisition module 610, a score determination module 620, an order determination module 630, and a content pushing module 640.

[0164] The record acquisition module 610 is configured to acquire a viewing record of a target object on first media content, the viewing record being used to reflect a viewing preference of the target object on second media content; wherein the first media content and the second media content belong to two different domains.

[0165] The score determination module 620 is configured to determine, according to the viewing record, a recommendation score of n items of second media content in each recommendation strategy respectively, n being a positive integer.

[0166] The order determination module 630 is configured to determine a push order of the n items of second media content based on the recommendation scores of the n items of second media content in each of the recommendation strategies respectively.

[0167] The content pushing module 640 is configured to push the second media content to the target object according to the push order.

[0168] In summary, the technical solutions provided by the embodiments of the present application determine a push order of pushing second media content to a target object according to a viewing record of the target object on first media content, the viewing record being used to reflect a viewing preference of the target object on the second media content, thereby migrating the behavior portrait of the target object in one domain to another domain, realizing cross-domain pushing of the second media content that the target object is more interested in to the target object, and improving the accuracy of media content pushing at cold start.

[0169] ​In some embodiments, the score determining module 620 is configured to:

[0170] obtain an association weight between each category of the second media content and each category of the first media content, the association weight being used to indicate an association degree between a category of the second media content and a category of the first media content;

[0171] determine, according to the viewing record and the association weight, a first recommendation score of each category of the second media content corresponding to the target object, the first recommendation score being used to indicate a degree of attention of the target object to a category to which the second media content belongs;

[0172] determine a first recommendation score of the n pieces of second media content based on the first recommendation score of each category of the second media content corresponding to the target object; wherein a first recommendation score of an i-th piece of second media content in the n pieces of second media content is a first recommendation score of a category to which the i-th piece of second media content belongs corresponding to the target object, and i is a positive integer.

[0173] In some embodiments, as shown in Figure 7 the score determining module 620 further includes a record obtaining submodule 621, a target value determining submodule 622, and a score determining submodule 623.

[0174] The record obtaining submodule 621 is configured to obtain, from the viewing record, a viewing record of the target object for a popular first media content, the popular first media content being a first media content meeting a first popularity condition.

[0175] The target value determining submodule 622 is configured to determine a target value according to the viewing record of the target object for the popular first media content, the target value being a recommendation score of a popular second media content corresponding to the target object, the popular second media content being a second media content meeting a second popularity condition.

[0176] The score determining submodule 623 is configured to, for an i-th piece of second media content in the n pieces of second media content, if the i-th piece of second media content belongs to the popular second media content, determine a second recommendation score of the i-th piece of second media content as the target value, and i is a positive integer.

[0177] In some embodiments, the first media content is a film and television content, the popular first media content includes a popular film and a popular episode; the second media content is a news content, and the popular second media content includes a popular news; as shown in Figure 7 the target value determining submodule 622 is configured to:

[0178] obtaining a first score based on a record of watching the popular movie in the view record for the popular first media content;

[0179] obtaining a second score based on an average duration from starting to watching each episode of the popular series to watching the end of the episode and a number of the popular series watched in a first preset time period in the view record for the popular first media content;

[0180] determining the target value according to the first score and the second score.

[0181] In some embodiments, as shown in FIG. 6, the score determining module 620 further includes a weight determining submodule 624. Figure 7

[0182] The weight determining submodule 624 is configured to determine a view similarity weight of each first object other than the target object based on the view record, the view similarity weight of the mth first object being used to indicate a similarity between the view record of the mth first object and the view record relative to other first objects, m being a positive integer.

[0183] The record obtaining submodule 621 is further configured to obtain a view record of each first object for the second media content.

[0184] The score determining submodule 623 is further configured to determine a third recommendation score of each second media content according to the view similarity weight and the view record of each first object for the second media content.

[0185] In some embodiments, as shown in FIG. 6, the weight determining submodule 624 is configured to: Figure 7

[0186] calculate a view similarity between the view record of each first object and the view record;

[0187] determine a view similarity weight corresponding to each first object based on the view similarity corresponding to each first object.

[0188] In some embodiments, the score determining module 620 is configured to:

[0189] For the ith second media content in the n second media contents, if the ith second media content is real-time second media content, a fourth recommendation score of the ith second media content is determined as a first value, the real-time second media content being second media content generated in a second preset time period.

[0190] ​​Or, for the i th second media content of the n second media contents, if the i th second media content belongs to the relevant second media content, the fourth recommendation score of the i th second media content is a second value, the relevant second media content is the second media content related to the first media content contained in the viewing record; wherein, the i is a positive integer.

[0191] In some embodiments, the sorting determining module 630 is configured to:

[0192] For the n second media contents, the maximum value of the recommendation scores of the i th second media content in each of the push strategies is determined as the final recommendation score of the i th second media content, and the i is a positive integer.

[0193] The push sorting is determined according to the n second media contents respectively corresponding to the latest recommendation scores.

[0194] In some embodiments, the sorting determining module 630 is configured to:

[0195] The priority of each of the recommendation strategies is obtained.

[0196] The push sorting is determined according to the n second media contents respectively in each of the push strategies and the priority of each of the recommendation strategies.

[0197] In some embodiments, the sorting determining module 630 is configured to:

[0198] The recommendation weight corresponding to each of the recommendation strategies is obtained.

[0199] The final recommendation score corresponding to each of the n second media contents is determined according to the recommendation score of each of the n second media contents in each of the push strategies and the recommendation weight corresponding to each of the recommendation strategies.

[0200] The push sorting is determined according to the n second media contents respectively corresponding to the final recommendation scores.

[0201] It should be noted that the apparatus provided in the above embodiments, in realizing its functions, only takes the above-mentioned division of each functional module as an example, and in actual application, the above-mentioned functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above-described functions. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is described in detail in the method embodiments, which will not be repeated here.

[0202] Please refer to Figure 8, which shows a structural block diagram of a computer device 800 provided in one embodiment of the present application. The computer device 800 may be the terminal or server described above, and is used to implement the media content push method provided in the above embodiment.

[0203] Typically, the computer device 800 includes a processor 801 and a memory 802 .

[0204] The processor 801 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 801 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field Programmable Gate Array), or PLA (Programmable Logic Array). The processor 801 may also include a main processor and a coprocessor. The main processor is a processor for processing data in an awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 801 may be integrated with a GPU (Graphics Processing Unit) or other processors, which is not limited in the embodiments of the present application.

[0205] The memory 802 may include one or more computer-readable storage media, which may be non-transitory. The memory 802 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices and flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 802 is used to store a computer program, and the computer program is configured to be executed by one or more processors to implement the above-mentioned media content push method.

[0206] In some embodiments, computer device 800 may optionally include a peripheral device interface 803 and at least one peripheral device. Processor 801, memory 802, and peripheral device interface 803 may be connected via a bus or signal lines. Each peripheral device may be connected to peripheral device interface 803 via a bus, signal lines, or circuit boards.

[0207] Those skilled in the art will understand that Figure 8 The structure shown in the figure does not constitute a limitation on the computer device 800, and the computer device 800 may include more or fewer components than shown in the figure, or combine some components, or adopt a different arrangement of components.

[0208] In the example embodiment, a computer readable storage medium is also provided, in which at least one instruction, at least one program, a code set or an instruction set is stored, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the above-mentioned media content pushing method.

[0209] Optionally, the computer readable storage medium can include a ROM (Read-Only Memory), a RAM (Random-Access Memory), a SSD (Solid State Drives), an optical disc, or the like. Among them, the random access memory can include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0210] In the example embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to make the computer device execute the above-mentioned media content pushing method.

[0211] It should be understood that "multiple" mentioned herein refers to two or more. The "and / or" describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the three cases of A existing alone, A and B existing together, and B existing alone. The character " / " generally represents that the associated objects before and after it are in an "or" relationship.

[0212] The above only describes the example embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A media content pushing method, characterized by, The method comprises: obtaining a viewing record of a target object on first media content, the viewing record being used to reflect a viewing preference of the target object on second media content; wherein the first media content and the second media content belong to two different fields; obtaining an association weight between each category of the second media content and each category of the first media content, the association weight being used to indicate an association degree between the category of the second media content and the category of the first media content; determining a first recommendation score of each category of the second media content according to the viewing record and the association weight, the first recommendation score being used to indicate a degree of attention of the target object to the category to which the second media content belongs; determining a first recommendation score of n items of second media content based on the first recommendation score of each category of the second media content; the first recommendation score of the i-th item of second media content is the first recommendation score of the category to which the i-th item of second media content belongs, the i being a positive integer, and the n being a positive integer; determining a push order of the n items of second media content based on a recommendation score of each of the n items of second media content in each of the recommendation strategies, the recommendation score in each of the recommendation strategies including the first recommendation score; pushing the second media content to the target object according to the push order.

2. The method of claim 1, wherein, The method further comprises: obtaining a viewing record of the target object on popular first media content from the viewing record, the popular first media content being first media content meeting a first popularity condition; determining a target value according to the viewing record of the target object on popular first media content, the target value being a recommendation score of popular second media content corresponding to the target object, the popular second media content being second media content meeting a second popularity condition; for the i-th item of second media content in the n items of second media content, if the i-th item of second media content belongs to the popular second media content, determining a second recommendation score of the i-th item of second media content as the target value, the i being a positive integer, and the recommendation score in each of the recommendation strategies including the second recommendation score.

3. The method of claim 2, wherein, The first media content is video content, the popular first media content including popular movies and popular episodes; the second media content is news content, the popular second media content including popular news; The method further comprises: obtaining a first score based on a record of watching the popular movies in the viewing record on the popular first media content; determining a second score based on an average duration from the start of watching each episode of the popular episodes to the end of watching in the viewing record on the popular first media content, and a number of the popular episodes watched in a first preset time period; determining the target value according to the first score and the second score.

4. The method of claim 1, wherein, The method further comprises: Determine, through the viewing record, a viewing similarity weight of each first object other than the target object, wherein the viewing similarity weight of an mth first object indicates a similarity between the viewing record of the mth first object and the viewing record, relative to other first objects, m being a positive integer; Obtain a viewing record of each first object on the second media content; Determine a third recommendation score of each item of the second media content according to the viewing similarity weight and the viewing record of each first object on the second media content, the recommendation score in each recommendation strategy including the third recommendation score.

5. The method of claim 4, wherein, The method further comprises: Calculate a viewing similarity between the viewing record of each first object and the viewing record; Determine a viewing similarity weight corresponding to each first object based on the viewing similarity corresponding to each first object.

6. The method of claim 1, wherein, The method further comprises: For an ith item of the n items of second media content, if the ith item of second media content belongs to real-time second media content, determine a fourth recommendation score of the ith item of second media content as a first numerical value, wherein the real-time second media content is second media content generated within a second preset time period; Or, For an ith item of the n items of second media content, if the ith item of second media content belongs to related second media content, determine a fourth recommendation score of the ith item of second media content as a second numerical value, the related second media content being second media content related to the first media content contained in the viewing record; Wherein, i is a positive integer, and the recommendation score in each recommendation strategy includes the fourth recommendation score.

7. The method according to any one of claims 1 to 6, characterized in that, The method further comprises: For the n items of second media content, determine a maximum value of the recommendation score in each push strategy of an ith item of second media content as a final recommendation score of the ith item of second media content, i being a positive integer; Determine the push order according to the latest recommendation score corresponding to the n items of second media content.

8. The method according to any one of claims 1 to 6, characterized in that, The method further comprises: Obtain a priority of each recommendation strategy; Determine the push order according to the recommendation score of the n items of second media content in each push strategy and the priority of each recommendation strategy.

9. The method according to any one of claims 1 to 6, characterized in that, The method further comprises: Obtain a recommendation weight corresponding to each recommendation strategy; determine, according to the recommendation scores of the n items of second media content in each of the recommendation strategies respectively and the recommendation weights corresponding to each of the recommendation strategies respectively, final recommendation scores corresponding to the n items of second media content respectively; determine the push order according to the final recommendation scores corresponding to the n items of second media content respectively.

10. A media content pushing apparatus, characterized by, The apparatus comprises: a record obtaining module configured to obtain a viewing record of a target object on first media content, the viewing record being used to reflect a viewing preference of the target object on second media content; wherein the first media content and the second media content belong to two different fields; a score determining module configured to obtain an association weight between each category of the second media content and each category of the first media content, the association weight being used to indicate an association degree between the category of the second media content and the category of the first media content; the score determining module is further configured to determine a first recommendation score of each category of the second media content according to the viewing record and the association weight, the first recommendation score being used to indicate a degree of attention of the target object to the category to which the second media content belongs; the score determining module is further configured to determine a first recommendation score of n items of second media content based on the first recommendation scores of each category of the second media content; a first recommendation score of an i-th item of second media content is a first recommendation score of a category to which the i-th item of second media content belongs, the i being a positive integer, and the n being a positive integer; a sorting determining module configured to determine a push order of the n items of second media content based on recommendation scores of the n items of second media content in each of the recommendation strategies respectively, the recommendation scores in each of the recommendation strategies including the first recommendation scores; a content push module configured to push the second media content to the target object according to the push order.

11. The apparatus of claim 10, wherein, The score determining module comprises: a record obtaining sub-module configured to obtain, from the viewing record, a viewing record of the target object on a popular first media content, the popular first media content being first media content meeting a first popularity condition; a target value determining sub-module configured to determine a target value according to the viewing record of the target object on the popular first media content, the target value being a recommendation score of a popular second media content corresponding to the target object, the popular second media content being second media content meeting a second popularity condition; a score determining sub-module configured to, for an i-th item of second media content of the n items of second media content, if the i-th item of second media content belongs to the popular second media content, determine a second recommendation score of the i-th item of second media content as the target value, the i being a positive integer, and the recommendation scores in each of the recommendation strategies including the second recommendation scores.

12. The apparatus of claim 11, wherein, The first media content is a video content, the popular first media content includes a popular movie and a popular episode; the second media content is a news content, the popular second media content includes a popular news; the target value determining sub-module is configured to: obtain a first score based on a record of watching the popular movie in a view record of the popular first media content; obtain a second score based on an average duration from starting to watching each episode of the popular series in the view record of the popular first media content and a number of the popular series watched in a first preset time period; determine the target value according to the first score and the second score.

13. The apparatus of claim 10, wherein, The score determination module comprises: The weight determination submodule is configured to determine a view similarity weight of each first object other than the target object based on the view record, wherein the view similarity weight of an mth first object is used to indicate a similarity between the view record of the mth first object and the view record relative to other first objects, and m is a positive integer. The record acquisition submodule is further configured to acquire a view record of the second media content of each first object. The score determination submodule is further configured to determine a third recommendation score of each item of the second media content according to the view similarity weight and the view record of each first object for the second media content, and the recommendation score in each recommendation strategy comprises the third recommendation score.

14. The apparatus of claim 13, wherein, The weight determination submodule is configured to: calculate a view similarity between the view record of each first object and the view record; determine a view similarity weight corresponding to each first object based on the view similarity corresponding to each first object.

15. The apparatus of claim 10, wherein, The score determination submodule is further configured to: for an ith item of the n items of the second media content, if the ith item of the second media content belongs to real-time second media content, determine a fourth recommendation score of the ith item of the second media content as a first numerical value, wherein the real-time second media content is second media content generated within a second preset time period; or, for an ith item of the n items of the second media content, if the ith item of the second media content belongs to related second media content, determine a fourth recommendation score of the ith item of the second media content as a second numerical value, wherein the related second media content is second media content related to the first media content contained in the view record; wherein i is a positive integer, and the recommendation score in each recommendation strategy comprises the fourth recommendation score.

16. The apparatus of any one of claims 10 to 15, wherein, The sorting determination module is configured to: for the n items of the second media content, determine a maximum value of the recommendation score in each push strategy as a final recommendation score of an ith item of the second media content, and i is a positive integer; determine the push sorting according to the latest recommendation score corresponding to the n items of the second media content.

17. The apparatus of any one of claims 10 to 15, wherein, The sorting determination module is configured to: acquire a priority of each recommendation strategy; determine the push sorting according to the recommendation score of each item of the n items of the second media content in each push strategy and the priority of each recommendation strategy.

18. The apparatus of any one of claims 10 to 15, wherein, The sorting determination module is configured to: acquire a recommendation weight corresponding to each recommendation strategy; According to the recommendation scores of the n items of second media content in each of the push strategies respectively, and the recommendation weights corresponding to each of the recommendation strategies respectively, final recommendation scores corresponding to the n items of second media content are determined respectively; According to the final recommendation scores corresponding to the n items of second media content respectively, the push order is determined.

19. A computer device, comprising: The computer device comprises a processor and a memory, the memory stores a computer program, the computer program is loaded and executed by the processor to implement the media content push method in any one of claims 1 to 9.

20. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, the computer program is loaded and executed by the processor to implement the media content push method in any one of claims 1 to 9.

21. A computer program product, characterised in that, The computer program product comprises computer instructions stored in a computer readable storage medium, the processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the media content push method in any one of claims 1 to 9.

Images