Media account searching method and device, equipment and storage medium

Abstract

The invention provides a media account searching method and device, equipment and a storage medium, can be applied to the technical fields of self-media, social media, artificial intelligence and the like, and comprises the following steps: obtaining N media contents of a first media account, and determining label data of the ith media content under K types of different labels based on content information of the ith media content; determining a popularity value of the i-th media content based on the operation data of the Q object operations of the i-th media content; and determining label data of the first media account under the K types of different labels based on the popularity value of each media content in the N media contents and the label data under the K types of different labels, and searching a target media account based on the label data of the first media account under the K types of different labels. According to the method, the tag data of the account dimension is determined based on the media content of the first media account so as to expand the account tag, and the search accuracy of the media account is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, and in particular to a media account search method, apparatus, device, and storage medium. Background Art

[0002] With the rapid development of internet technology, various media platforms have emerged, allowing users to publish media content on these platforms. In some cases, users need to search for a target media account among a multitude of media accounts hosting this content. Currently, many media platforms offer a media account search function. This means that being able to quickly and accurately search for a media account based on user needs is a crucial capability for these platforms.

[0003] Currently, when searching for media accounts, information is typically extracted based on the blogger's media account name (nickname) and media account profile. This information is then segmented and indexed to create an index. Intersection hits are then performed based on this index to retrieve media accounts. However, current searches based on media account names and profiles are incapable of accurately searching for media accounts. Summary of the Invention

[0004] The present application provides a media account search method, apparatus, device, and storage medium for improving the search accuracy of media accounts.

[0005] In a first aspect, the present application provides a method for identifying an abnormal subject, comprising:

[0006] Based on the search request for the target media account, obtain N media content published by a first media account, where the first media account is any candidate media account of the target media account, and N is a positive integer;

[0007] For an i-th media content among the N media content, determining label data of the i-th media content under K different labels based on content information of the i-th media content, where i is a positive integer less than or equal to N, and K is a positive integer;

[0008] Determine a popularity value of the i-th media content based on operation data of Q types of object operations on the i-th media content, where Q is a positive integer;

[0009] Based on the popularity value of each media content in the N media content and the tag data under the K different tags, the tag data of the first media account under the K different tags are determined, and based on the tag data of the first media account under the K different tags, the target media account search is performed.

[0010] In a second aspect, the present application provides a media account search device, comprising:

[0011] an acquiring unit, configured to acquire, based on a search request of a target media account, N media contents published by a first media account, where the first media account is any candidate media account of the target media account, and N is a positive integer;

[0012] a content label determination unit, configured to determine, for an i-th media content among the N media content, label data of the i-th media content under K different labels based on content information of the i-th media content, where i is a positive integer less than or equal to N, and K is a positive integer;

[0013] a popularity determining unit, configured to determine a popularity value of the i-th media content based on operation data of Q types of object operations on the i-th media content, where Q is a positive integer;

[0014] An account label determination unit is used to determine the label data of the first media account under the K different labels based on the popularity value of each media content in the N media content and the label data under the K different labels, and to search for the target media account based on the label data of the first media account under the K different labels.

[0015] In some embodiments, the content label determination unit is specifically used to determine, for the j-th label among the K different labels, based on the content information of the i-th media content, a probability value that the i-th media content belongs to each of the P subcategories under the j-th label, where j is a positive integer less than or equal to K, and P is a positive integer; and based on the probability value that the i-th media content belongs to each of the P subcategories under the j-th label, determine the label data of the i-th media content under the j-th label.

[0016] In some embodiments, the content label determination unit is specifically configured to classify the content information of the i-th media content using a classification model corresponding to the j-th label to obtain a probability value of the i-th media content belonging to each of the P subcategories under the j-th label.

[0017] In some embodiments, the content tag determination unit is specifically used to obtain M media content published by the first media account, where M is a positive integer greater than 1; based on the operation data of Q types of object operations on the M media content, determine the weights corresponding to the Q types of object operations; based on the weights corresponding to the Q types of object operations and the operation data of the Q types of object operations on the i-th media content, determine the popularity value of the i-th media content.

[0018] In some embodiments, the content tag determination unit is specifically used to select R media data from the M media content based on the number of clicks on each media content in the M media content, where R is a positive integer less than or equal to M; and determine the weights corresponding to the Q object operations based on the operation data of the Q object operations of the R media content.

[0019] In some embodiments, the content label determination unit is specifically used to determine, for the kth object operation among the Q object operations, a first number of media contents whose operation data of the kth object operation is greater than a corresponding threshold from the R media content, where k is a positive integer less than or equal to Q; and determine the weight corresponding to the kth object operation based on the first number corresponding to the kth object operation and the number of the R media content.

[0020] In some embodiments, the content label determination unit is specifically configured to determine a ratio of a first quantity corresponding to the k-th object operation to the quantity of the R media contents as a weight corresponding to the k-th object operation.

[0021] In some embodiments, the heat determination unit is specifically used to determine the second numerical value corresponding to the kth object operation of the i-th media content based on the operation data of the kth object operation of the i-th media content and the threshold value corresponding to the kth object operation; determine the heat value of the kth object operation of the i-th media content based on the second numerical value corresponding to the kth object operation of the i-th media content and the weight corresponding to the kth object operation; determine the heat value of the i-th media content based on the heat values ​​of the Q types of object operations of the i-th media content.

[0022] In some embodiments, the heat determination unit is specifically used to determine that the second value is a first preset value if the operation data of the kth object operation of the ith media content is greater than or equal to the threshold corresponding to the kth object operation; if the operation data of the kth object operation of the ith media content is less than the threshold corresponding to the kth object operation, then determine that the second value is a second preset value, and the second preset value is less than the first preset value.

[0023] In some embodiments, the heat determination unit is specifically used to determine the product of the second numerical value corresponding to the kth object operation of the i-th media content and the weight corresponding to the kth object operation as the heat value of the kth object operation of the i-th media content.

[0024] In some embodiments, the popularity determining unit is specifically configured to determine the sum of popularity values ​​of Q types of object operations on the i-th media content as the popularity value of the i-th media content.

[0025] In some embodiments, if the label data of the i-th media content under the j-th label includes the probability value of the i-th media content belonging to each of the P subcategories under the j-th label, the account label determination unit is specifically used to determine the score of the first media account in each of the P subcategories under the j-th label based on the probability value of each of the P subcategories under the j-th label of the N media content and the popularity value of the N media content; and determine the label data of the first media account under the K different labels based on the score of the first media account in each subcategories under the K different labels.

[0026] In some embodiments, the account label determination unit is specifically used to determine the score of the first media account in the nth subcategory among the P subcategories based on the probability values ​​of the N media contents respectively belonging to the nth subcategory and the popularity values ​​of the N media contents, where n is a positive integer less than or equal to P; and determine the label data of the first media account under the jth category label based on the scores of the first media account in the P subcategories in the jth category label.

[0027] In some embodiments, the account label determination unit is specifically used to select T media content from the N media contents based on the probability values ​​that the N media content respectively belong to the nth subcategory, where T is a positive integer less than or equal to N; and determine the score of the first media account in the nth subcategory based on the probability values ​​that the T media content respectively belong to the nth subcategory and the popularity values ​​of the T media content.

[0028] In some embodiments, the account label determination unit is specifically used to multiply the probability value of each of the T media contents that the media content belongs to the nth subcategory by the popularity value of the media content to obtain a third value; and add the third values ​​corresponding to each of the T media contents to obtain the score of the first media account in the nth subcategory.

[0029] In some embodiments, the account label determination unit is specifically used to determine the sum of the scores of the first media account in the P subcategories under the j-th category label; for the n-th subcategories, based on the ratio of the score of the first media account in the n-th subcategories to the sum of the scores, determine the proportion of the n-th subcategories; based on the proportion of the first media account in each of the P subcategories, select at least one subcategories from the P subcategories and determine it as the label data of the first media account under the j-th category label.

[0030] In some embodiments, the account label determination unit is specifically used to, in some embodiments, the account label determination unit 14 is specifically used to select the at least one subcategory from the P subcategories based on the score of the first media account in each of the P subcategories and the proportion of each subcategory, and determine it as the label data of the first media account under the j-th category label.

[0031] In a third aspect, an electronic device is provided, comprising a processor and a memory, wherein the memory is configured to store a computer program, and the processor is configured to call and execute the computer program stored in the memory to execute the method of the first aspect and its implementations.

[0032] In a fourth aspect, a chip is provided for implementing the method described in any one of the first aspects and its implementations. Specifically, the chip includes a processor configured to retrieve and execute a computer program from a memory, causing a device equipped with the chip to execute the method described in any one of the first aspects and its implementations.

[0033] In a fifth aspect, a computer-readable storage medium is provided for storing a computer program, wherein the computer program enables a computer to execute the method in the above-mentioned first aspect and its various implementations.

[0034] In a sixth aspect, a computer program product is provided, comprising computer program instructions, wherein the computer program instructions enable a computer to execute the method in the above-mentioned first aspect and its various implementations.

[0035] In a seventh aspect, a computer program is provided, which, when executed on a computer, enables the computer to execute the method in the above-mentioned first aspect and its various implementations.

[0036] In summary, when searching for a target media account, the present application obtains N media contents published by the first media account for any candidate media account corresponding to the target media account, such as the first media account. For the i-th media content in the N media contents, the label data of the i-th media content under K different labels is determined based on the content information of the i-th media content. At the same time, the popularity value of the i-th media content is determined based on the operation data of the Q types of object operations of the i-th media content. Finally, based on the popularity value of each media content in the N media contents and the label data of each media content under K different labels, the label data of the first media account under K different labels is determined, and then the target media account search is performed based on the label data of the first media account under K different labels. In other words, the embodiment of the present application determines the label data of the media content dimension based on the media content of the first media account, and then determines the label data of the account dimension based on the popularity value of the media content and the label data of the media content dimension, so as to expand the label of the first media account, which can improve the search accuracy and search capability of the media account. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 A schematic diagram of an implementation environment involved in an embodiment of the present application;

[0039] Figure 2 A flowchart of a method for searching for media accounts provided in one embodiment of the present application;

[0040] Figure 3 A schematic diagram showing the probability values ​​of each sub-category predicted for the i-th media content;

[0041] Figure 4 A schematic diagram showing the probability of predicting the i-th media content belonging to each subcategory under the entity label;

[0042] Figure 5 A flowchart of a method for searching for media accounts provided in one embodiment of the present application;

[0043] Figure 6 is a schematic block diagram of a media account search device provided in one embodiment of the present application;

[0044] Figure 7 It is a schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0046] It should be noted that the terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In an embodiment of the present invention, "B corresponding to A" means that B is associated with A. In one implementation, B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B solely based on A, but that B can also be determined based on A and / or other information. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or devices. In the description of this application, unless otherwise specified, "plurality" means two or more than two.

[0047] The technical solution proposed in this application can be applied to technical fields such as self-media, artificial intelligence, social media, and new media to achieve accurate searches for media accounts.

[0048] The following is an introduction to the relevant concepts involved in the embodiments of this application.

[0049] Short videos, also known as short clips, are a form of internet content dissemination, typically consisting of videos lasting less than n minutes and distributed on new internet media. With the widespread adoption of mobile devices and increased network speeds, short, fast, and high-volume content is gaining popularity. In some embodiments, the short videos in the present application can refer to any short video platform product.

[0050] Artificial Intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also involves studying the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0051] Artificial intelligence (AI) technology is a comprehensive discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0052] Machine learning (ML) is a multidisciplinary field that encompasses probability theory, statistics, approximation theory, convex analysis, and algorithmic complexity theory. It specifically studies how computers can simulate or implement human learning behaviors to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is at the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications span all areas of AI. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and self-learning.

[0053] With the research and advancement of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, autonomous driving, drones, robots, smart medical care, smart customer service, etc. It is believed that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0054] In an embodiment of the present application, artificial intelligence technology is applied to media account search to achieve fast search of media accounts.

[0055] Currently, when searching for media accounts, information is usually extracted based on the blogger's media account name (nickname) and media account profile, and an index is constructed based on these two types of information. Then, the media account is recalled by intersection hit based on the index. However, most accounts and account profiles contain relatively little information, and sometimes include some useless information, so it is relatively difficult to extract information directly from the basic information of the account. For some content-related needs, this type of literal matching-based method cannot achieve accurate account search. For example, the search term is "BC+Funny", and the search system is expected to search for the ABC account. However, since the ABC account profile does not include the funny tag, the two words funny cannot be matched when searching based on the account name and account profile, and thus the ABC account cannot be accurately searched. It can be seen from this that the current search for media accounts based on media account names and media account profiles has the problem of inaccurate search.

[0056] In order to solve the above technical problems, when searching for a target media account, the embodiment of the present application obtains N media contents published by the first media account for any candidate media account corresponding to the target media account, such as the first media account, and for the i-th media content among the N media contents, determines the label data of the i-th media content under K different labels based on the content information of the i-th media content. At the same time, determines the popularity value of the i-th media content based on the operation data of Q types of object operations of the i-th media content. Finally, based on the popularity value of each media content in the N media contents and the label data of each media content under K different labels, determines the label data of the first media account under K different labels, and then performs a target media account search based on the label data of the first media account under K different labels. That is to say, the embodiment of the present application determines the label data of the media content dimension based on the media content of the first media account, and then determines the label data of the account dimension based on the popularity value of the media content and the label data of the media content dimension. When the label data of the account dimension is used as the supplementary label data of the first media account for account search, the accuracy of the media account search can be improved.

[0057] The following introduces the implementation environment of the media account search method provided in the embodiment of the present application.

[0058] Figure 1 1 is a schematic diagram of an implementation environment involved in an embodiment of the present application, including a terminal device 101 and a server 102. In the embodiment of the present application, the terminal device 101 is installed with a front end of a media platform, and the server 102 can be understood as a back end of the media platform.

[0059] In some embodiments, the server 102 of the present application includes classification models corresponding to different class labels, for example, a classification model corresponding to a category label and / or a classification model corresponding to an entity label. The classification model corresponding to the category label can predict the probability that the media content belongs to different category names, and the classification model corresponding to the entity label can predict the probability that the media content belongs to different entity names.

[0060] In the embodiments of this application, Figure 1 As shown, a user enters a search request for a target media account in the media platform of terminal device 101. Terminal device 101 sends the search request to server 102. Based on the search request, server 102 retrieves N media content published by any candidate media account for the target media account, such as the first media account. For the i-th media content in these N media content, server 102 determines the tag data for K different tags based on the content information of the i-th media content. Simultaneously, server 102 determines the popularity value of the i-th media content based on the Q types of object operations on the i-th media content. Finally, server 102 determines the tag data for the first media account under K different tags based on the popularity value of each media content in the N media content and the tag data for each media content under K different tags. Furthermore, server 102 searches for the target media account based on the tag data for the first media account under K different tags. Finally, server 102 sends the relevant media content of the searched target media account to first terminal device 101 for display. That is to say, the embodiment of the present application determines the label data of the media content dimension based on the media content of the first media account, and then determines the label data of the account dimension based on the popularity value of the media content and the label data of the media content dimension. When the label data of the account dimension is used as the supplementary label data of the first media account for account search, the accuracy of the media account search can be improved.

[0061] The present embodiment of the present application does not limit the specific type of terminal device 101. In some embodiments, terminal device 101 may include, but is not limited to, mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, wearable smart devices, medical devices, and the like. Devices are often equipped with a display device, which can be a monitor, display screen, touch screen, and the like. The touch screen can also be a touch screen, touch panel, and the like.

[0062] In some embodiments, the server 102 may be one or more servers. When there are multiple servers, there are at least two servers for providing different services, and / or there are at least two servers for providing the same service, such as providing the same service in a load balancing manner, which is not limited in this embodiment of the present application. The above-mentioned server may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The server can also become a node of the blockchain.

[0063] In the embodiment of the present application, the terminal device 101 and the server 102 can be directly or indirectly connected through wired communication or wireless communication, and the present application does not impose any restrictions on this.

[0064] It should be noted that the implementation environment of the embodiment of the present application includes but is not limited to Figure 1 shown.

[0065] The following describes the technical solutions of the embodiments of the present application in detail through some embodiments. The following embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0066] Figure 2 A flowchart of a method for searching a media account provided in an embodiment of the present application. The execution subject of the embodiment of the present application is a device with a search function, such as a media account search device, referred to as a search device. In some embodiments, the search device can be Figure 1 The server in , or Figure 1 The terminal device in Figure 1 For ease of description, the embodiment of the present application is described by taking the execution subject as an electronic device as an example.

[0067] like Figure 2 As shown, the media account search process of the embodiment of the present application includes:

[0068] S101 : Based on a search request of a target media account, obtain N media contents published by a first media account.

[0069] The first media account is any candidate media account of the target media account, and N is a positive integer.

[0070] In an embodiment of the present application, the target media account to be searched can be any media account registered on a media platform. For example, the target media account can be a media account registered on a live broadcast platform, or a media account registered on a short video platform, or a media account registered on a long video platform, or a media account registered on an audio platform, or a media account registered on a social media platform, or a media account registered on a self-media publishing platform, or a media account registered on a graphic platform, or a media account registered on an audio and video platform, or a media account registered on a social platform, etc.

[0071] The search request for the target media account in the embodiment of the present application may include basic account information of the target media account to be searched, such as the media account name and other information.

[0072] In some embodiments, the search request may include not only the basic account information of the target media account but also characteristic information of the media content published by the target media account. For example, a media account named "ABC" often publishes funny media content. If an object (e.g., a user) wants to search for the media account "ABC", they enter the search request "BC+funny" in the search box of the media platform.

[0073] However, current media account search methods rely on the name and profile of a media account, without considering the media content, resulting in inaccurate searches. For example, current media account search methods cannot find the media account named "ABC" based on the search query "BC+funny".

[0074] In order to accurately search for media accounts, the embodiment of the present application considers the media content published by the media account when searching for the media account. Specifically, when searching for the target media account, N media content published by the first media account is obtained.

[0075] Among them, the first media account is any candidate media account of the target media account. Exemplarily, during the search process of the target media account, the electronic device searches for the target media account from multiple candidate media accounts. At this time, the first media account can be understood as any media account among these multiple candidate media accounts. For example, if the target media account is a media account of a short video platform, all or multiple media accounts on the short video platform can be determined as candidate media accounts of the target media account, and the first media account can be any media account among all or multiple media accounts on the short video platform. For another example, if the target media account is a media account of a self-media publishing platform, all or multiple media accounts on the self-media publishing platform can be determined as candidate media accounts of the target media account, and the first media account can be any media account among all or multiple media accounts on the self-media publishing platform. That is to say, in the embodiment of the present application, the processing process for each candidate media account of the multiple candidate media accounts of the target media account is basically the same. For the sake of convenience of description, a candidate media account (such as the first media account) is used as an example for explanation.

[0076] In some embodiments, the above-mentioned first media account can also be a candidate media account for a recall. For example, if the search request includes "BC+Funny", the electronic device first recalls multiple candidate media accounts whose names include "BC" from the media accounts included in the media platform based on the search word "BC", and records each of these multiple candidate media accounts as the first media account. That is to say, each of these multiple candidate media accounts is used as the first media account, and the method of the embodiment of the present application is executed to obtain the label data of these multiple candidate media accounts under K different labels, and then perform a target media account search based on the label data of these multiple candidate media accounts under K different labels. For example, based on the label data of these multiple candidate media accounts under K different labels and the search word "Funny" in the search request, the target media account is searched out from these multiple candidate media accounts.

[0077] In some embodiments, the first media account may also be a candidate media account among candidate media accounts that have been recalled multiple times based on other conditions, and this embodiment of the present application does not impose any limitation on this.

[0078] In some embodiments, the N media contents published by the first media account acquired by the electronic device may be the N media contents published by the first media account within a recent preset time period.

[0079] The embodiment of the present application does not limit the specific type of media content, for example, it can be at least one of video, article, and audio.

[0080] In the embodiment of the present application, after the electronic device obtains N media contents published by the first media account based on the search request of the target media account, it executes the following step S102.

[0081] S102 : For an i-th media content among N media content, determine label data of the i-th media content under K different labels based on content information of the i-th media content.

[0082] Wherein, i is a positive integer less than or equal to N, and K is a positive integer.

[0083] In the embodiment of the present application, after the electronic device obtains N media content published by the first media account based on the above steps, it determines, for each of the N media content, tag data for each media content under K different tags. In the embodiment of the present application, the specific process of the electronic device determining the tag data for each of the N media content under K different tags is basically the same. For ease of description, the i-th media content among the N media content is used as an example for explanation.

[0084] Specifically, for an i-th media content among N media content, the electronic device determines label data of the i-th media content under K different labels based on the content information of the i-th media content.

[0085] The K different types of labels mentioned above can be understood as labels from different perspectives of media content.

[0086] In some embodiments, the K different types of tags are K different types of preset tags. The embodiment of the present application does not limit the specific form of these K types of tags.

[0087] In an example, the K different tags may include at least one of a category tag, an entity tag, a style tag, and the like.

[0088] Among them, the category signature includes multiple category subcategories, such as life, entertainment, sports, music, games, movies, food, fashion, animation, etc.

[0089] The entity label includes multiple entity subcategories, such as people, places, objects, animals, etc.

[0090] The style tag includes multiple style subcategories, such as funny, happy, sad, etc.

[0091] In some embodiments, the above-mentioned K-category tags can also be finer-grained tags, such as secondary category tags, secondary entity tags, etc. For example, they can be finer-grained tags such as World of Warcraft-game, World of Warcraft-movie, etc.

[0092] It should be noted that the embodiment of the present application does not limit the specific forms of the K different types of tags and the sub-categories included in each tag, and can be set according to actual needs.

[0093] In the embodiment of the present application, for the i-th media content, the electronic device determines the tag data of the i-th media content under the K different tags based on the content information of the i-th media content.

[0094] The embodiment of the present application does not limit the specific manner in which the electronic device determines the tag data of the i-th media content under the K different tags based on the content information of the i-th media content.

[0095] In some embodiments, text information of the content information of the i-th media content is extracted, and label data that matches the K different types of labels are determined from the text information, and then the matching label data are determined as the label data of the i-th media content under the K different types of labels.

[0096] In some embodiments, the above S102 includes the following steps S102-A and S102-B:

[0097] S102-A. For a j-th label among K different labels, based on the content information of the i-th media content, determine a probability value that the i-th media content belongs to each of P subcategories under the j-th label, where j is a positive integer less than or equal to K, and P is a positive integer;

[0098] S102-B: Determine label data of the i-th media content under the j-th label based on the probability value of the i-th media content belonging to each of the P subcategories under the j-th label.

[0099] In this implementation, the specific method of determining the tag data of the i-th media content under each type of label in the K types of labels is basically the same. For ease of description, the j-th type of label is taken as an example for explanation.

[0100] Specifically, each of the K different labels includes at least one sub-category. For example, the jth label includes P sub-categories. It should be noted that the number of sub-categories included in different labels can be the same or different, and this embodiment of the application does not limit this. For example, the first label includes 10 sub-categories, and the second label includes 5 sub-categories.

[0101] In an embodiment of the present application, K different types of labels can be understood as different types of labels, and the sub-categories under each type of label can be understood as specific label data under that type of label. For example, for entity labels, the entity labels include: people, animals, objects, and other specific label categories. These specific label categories can be understood as sub-categories under the entity type label.

[0102] For example, assume that K class labels include category labels and entity labels, where the category label includes at least one subcategory and the entity label also includes at least one subcategory. For example, the category label includes n1 subcategories, that is, it includes n1 different category names, such as life, entertainment, sports, music, and food. For another example, the entity label includes n2 subcategories, that is, it includes n2 different entity names, such as entity names of people, places, objects, and animals.

[0103] In one example, the category label includes n1 subcategories, as shown in Table 1:

[0104] Table 1

[0105]

[0106] As shown in Table 1, the category tag includes n1 subcategories, namely cate1, cate2, cate3, ...caten1. Based on the content information of the i-th media content, the electronic device determines the probability value of the i-th media content belonging to the n1 subcategories of cate1, cate2, cate3, ...caten1 under the category tag.

[0107] For example, the probability values ​​of the i-th media content belonging to the target signature n1 subcategories are shown in Table 2:

[0108] Table 2

[0109]

[0110] As shown in Table 2, in an embodiment of the present application, the content information of the i-th media content is analyzed, and the probability value of the i-th media content belonging to the cate1 subclassification in the category tag is S11, the probability value of the i-th media content belonging to the cate2 subclassification in the category tag is S12, and the probability value of the i-th media content belonging to the cate3 subclassification in the category tag is S13.

[0111] In one example, the entity tag includes n2 subcategories, as shown in Table 3:

[0112] Table 3

[0113]

[0114] As shown in Table 3, the entity tag includes n2 subcategories, namely entity1, entity2, entity3, ..., En2. Based on the content information of the i-th media content, the electronic device determines the probability values ​​of the i-th media content belonging to the n2 subcategories of entity1, entity2, entity3, ..., En2 under the entity tag.

[0115] For example, the probability values ​​of the i-th media content belonging to n2 subcategories under the entity label are shown in Table 4:

[0116] Table 4

[0117]

[0118]

[0119] As shown in Table 4, in an embodiment of the present application, the content information of the i-th media content is analyzed, and the probability value of the i-th media content belonging to the entity1 subcategory in the entity tag is S21, the probability value of the i-th media content belonging to the entity2 subcategory in the entity tag is S22, and the probability value of the i-th media content belonging to the entity3 subcategory in the entity tag is S23.

[0120] The embodiment of the present application does not limit the specific manner in which the electronic device determines the probability value of the i-th media content belonging to each of the P subcategories under the j-th category label based on the content information of the i-th media content.

[0121] In one possible implementation, the electronic device extracts text information from the i-th media content. For example, if the i-th media content is a video, the electronic device extracts the audio data from the video and converts the audio data into text information. For another example, if the i-th media content is an image, the electronic device extracts the text information from the image as the text information for the i-th media content. Next, the electronic device matches the text information of the i-th media content with P subcategories, and based on the matching results, obtains a probability value that the i-th media content belongs to the P subcategories under the j-th label.

[0122] In one possible implementation, the electronic device may classify the content information of the i-th media content using a classification model corresponding to the j-th label to obtain a probability value of the i-th media content belonging to each of the P subcategories under the j-th label.

[0123] In this implementation, different class labels may correspond to different classification models.

[0124] In one example, assume that the j-th class label is the class label, such as Figure 3 As shown, the j-th category label corresponds to the first classification model, and the first classification model is a category extraction model, which is obtained by training with labeled category training samples and can predict the probability values ​​of the i-th media content belonging to each subcategory under the category label. For example, as shown in the formula categoryList = categoryModel(Mi), the i-th media content Mi is input into the first classification model categoryModel for category extraction, and the first classification model categoryModel outputs the probability values ​​of the i-th media content belonging to the three subcategories cate1, cate2 and cate3 under the category label, that is, categoryList = [(cate1, S11), (cate2, S12), (cate3, S13)], where cate1 is the category name 1 (i.e., subcategory 1), and S11 is the probability value of the i-th media content belonging to the category name cate1 (i.e., subcategory 1).

[0125] In one example, assume that the jth class label is an entity label, such as Figure 4 As shown, the j-th class label corresponds to the second classification model, and the second classification model is an entity extraction model, which is trained with labeled entity training samples and can predict the probability values ​​of the i-th media content belonging to each sub-category under the entity label. For example, as shown in the formula entityList = NERModel (doc), the i-th media content Mi is input into the second classification model entityList for processing, and the second classification model entityList1 outputs the probability values ​​of the i-th media content belonging to the three sub-categories entity1, entity2 and entity3 under the entity label, that is, entityList = [(entity1, S21), (entity2, S22), (entity3, S23)], where entity1 is the entity name 1 (i.e., sub-category 1), and S21 is the probability value of the i-th media content belonging to the entity name entity1 (i.e., sub-category 1).

[0126] Based on the above steps, the electronic device can determine the probability values ​​of the i-th media content among the N media content belonging to each subcategory under each category of the K category tags. For example, if the K category tags include target tags and entity tags, the electronic device can determine the probability values ​​of the i-th media content belonging to each subcategory within the target tag, and the probability values ​​of the i-th media content belonging to each subcategory within the entity tag.

[0127] For the j-th category label among the K category labels, the electronic device determines the probability value of the i-th media content belonging to each of the P subcategories under the j-th category label based on the above steps, and then determines the label data of the i-th media content under the j-th category label based on the probability value of the i-th media content belonging to each of the P subcategories under the j-th category label.

[0128] For example, the probability value of the i-th media content belonging to each of the P subcategories under the j-th category label is determined as the label data of the i-th media content under the j-th category label. In this case, the label data of the i-th media content under the j-th category label includes the probability value of the i-th media content belonging to each of the P subcategories.

[0129] For another example, based on the probability values ​​of the i-th media content belonging to each of the P subcategories under the j-th label, at least one subcategories is selected from the P subcategories as the label data for the i-th media content under the j-th label. For example, the subcategories corresponding to the largest probability value or values ​​among the P subcategories for the i-th media content belonging to are determined as the label data for the i-th media content under the j-th label.

[0130] The above describes the specific process of determining the tag data of the i-th media content under the j-th tag in the K-type tags. Based on the same method, the electronic device can determine the tag data of the i-th media content under each type of tag in the K-type tags.

[0131] S103 : Determine the popularity value of the i-th media content based on the operation data of the Q types of object operations on the i-th media content.

[0132] Where Q is a positive integer

[0133] It should be noted that there is no order of execution between the above S103 and the above S102. That is to say, the above S103 can be executed before the above S102, or after the above S102, or simultaneously with the above S102. The embodiment of the present application does not limit this.

[0134] In an embodiment of the present application, when determining the tag data for a first media account based on the tag data of media content published by the first media account, the popularity of the media content is taken into account, thereby enhancing the influence of the tag data of media content with high popularity on the tag data of the first media account, thereby improving the accuracy of determining the tag data for the first media account. When performing a subsequent search for a target media account based on the accurately determined tag data for the first media account, the accuracy of the search for the target media account can be improved.

[0135] The following describes the specific process of determining the popularity value of the media content of the first media account. In the embodiment of the present application, the specific process of determining the popularity value of each of the N media contents of the first media account is basically the same. For ease of description, the process of determining the popularity value of the i-th media content is used as an example for explanation.

[0136] In this embodiment of the present application, for the i-th media content, the popularity value of the i-th media content is determined based on the operation data of Q types of object operations on the i-th media content. The embodiment of the present application does not limit the specific types of the Q types of object operations, and for example, it may include any operation behavior of an object (such as a user) on the i-th media content.

[0137] Exemplarily, the Q types of object operations may include at least one operation behavior of the object such as liking, forwarding, reading, and collecting the i-th media content.

[0138] The operation data of an object operation can be understood as the number of times the object operation is performed on the i-th media content. For example, if the object operation is a like, the like operation data can be the number of likes. For another example, if the object operation is a forward, the forward operation data can be the number of forwards.

[0139] In one example, the Q types of object operations mentioned above may be preset.

[0140] The embodiment of the present application does not limit the specific manner in which the electronic device determines the popularity value of the i-th media content based on the operation data of Q types of object operations based on the i-th media content.

[0141] In some embodiments, the electronic device determines the popularity value of the i-th media content based only on the operation data of the Q types of object operations on the i-th media content. In one example, the heat value of the i-th media content can be determined by the sum of the operation data of the Q types of object operations on the i-th media content. For example, if the Q types of object operations include forwarding, reading, and collecting, the sum of the forwarding amount, reading amount, and collection amount of the i-th media content can be determined as the heat value of the i-th media content. In one example, the heat value of the i-th media content can be determined by the average value of the operation data of the Q types of object operations on the i-th media content. For example, if the Q types of object operations include forwarding, reading, and collecting, the average value of the forwarding amount, reading amount, and collection amount of the i-th media content can be determined as the heat value of the i-th media content.

[0142] In some embodiments, the above S103 includes the following steps S103-A to S103-C:

[0143] S103-A, obtaining M media contents published by the first media account, where M is a positive integer greater than 1;

[0144] S103-B, determining weights corresponding to the Q types of object operations based on the operation data of the Q types of object operations on the M media contents;

[0145] S103-C: Determine the popularity value of the i-th media content based on the weights corresponding to the Q types of object operations and the operation data of the Q types of object operations on the i-th media content.

[0146] In this implementation, when determining the popularity value of the i-th media content based on the operation data of Q types of object operations on the i-th media content, the weights of these Q types of object operations are first determined. The specific process of determining the weights of the Q types of object operations is described below.

[0147] Specifically, the electronic device randomly extracts M media contents published by the first media account from the online database. It should be noted that these M media contents may be consistent with the above-mentioned N media contents, or inconsistent with the above-mentioned N media contents, or consistent with part of the above-mentioned N media contents, and the embodiment of the present application does not limit this. Then, the electronic device determines the weights corresponding to the Q types of object operations based on the operation data of the Q types of object operations of the M media contents. For example, for each of the Q types of object operations, the weight corresponding to the object operation is determined based on the operation data of the object operation of the M media contents.

[0148] In some embodiments, in order to determine the accuracy of the weight of the extraction object operation, the randomly selected M media contents are first screened to select high-quality media contents for weight determination. In this case, the above S103-B includes the following steps:

[0149] S103-B1. Based on the number of clicks on each of the M media contents, select R media data from the M media contents, where R is a positive integer less than or equal to M;

[0150] S103-B2: Determine weights corresponding to the Q types of object operations based on the operation data of the Q types of object operations on the R media contents.

[0151] In an implementation, to improve the accuracy of weight determination, R high-quality media data are selected from the M media content based on the number of clicks on each of the M media content. For example, based on the number of clicks on each of the M media content, the R media data with the highest number of clicks are selected from the M media content. Furthermore, based on the operation data of the Q types of object operations on the R media data, the weights corresponding to the Q types of object operations are determined.

[0152] The embodiment of the present application does not limit the specific method of determining the weights corresponding to the Q types of object operations based on the operation data of the Q types of object operations of the R media data by the electronic device.

[0153] In one possible implementation, for each of the Q object operations, a weight is determined based on the operation data of the object operation in the R media contents. For example, the sum of the operation data of the object operation in the R media contents is determined as the weight of the object operation. For another example, a threshold value corresponding to the sum of the operation data of the object operation in the R media contents is determined as the weight of the object operation, where a larger sum value corresponds to a larger corresponding threshold value.

[0154] In a possible implementation, the electronic device determines the weight corresponding to the object operation based on the following steps S103-B21 and S103-B22:

[0155] S103-B21, for a k-th object operation among the Q object operations, determining a first number of media contents having operation data of the k-th object operation greater than a corresponding threshold from R media contents, where k is a positive integer less than or equal to Q;

[0156] S103-B22: Determine a weight corresponding to the k-th object operation based on the first number corresponding to the k-th object operation and the number of R media contents.

[0157] In this implementation, in the embodiment of the present application, the specific method of determining the weight of each object operation among the Q object operations is basically the same. For the sake of convenience of description, the specific process of determining the weight of the kth object operation among the Q object operations is explained as an example.

[0158] Specifically, for the kth object operation among the Q object operations, a first number of media contents having operation data of the kth object operation greater than a corresponding threshold is determined from the R media contents.

[0159] In an example, if the kth object operation is a forwarding operation and the threshold corresponding to the forwarding operation is a first threshold count_f, a first number count(forward) of media contents whose forwarding numbers are greater than the first threshold count_f is determined from the R media contents.

[0160] Exemplarily, the electronic device may determine a first number of media contents whose forwarding amount is greater than a first threshold from the R media contents based on the following formula (1):

[0161] count(forward) = sum(forward_i > count_f) (1)

[0162] Among them, count(forward) represents the number of media contents among the R media contents whose forwarding volume is greater than the first threshold, that is, the first number, count_f is the first threshold corresponding to the forwarding operation, and forward_i>count_f represents whether the forwarding volume of the i-th media content among the R media contents is greater than the first threshold. In this embodiment of the present application, if the forwarding volume of the media content is greater than the first threshold, the judgment result of forward_i>count_f is 1, otherwise the judgment result of forward_i>count_f is 0.

[0163] In an example, if the kth object operation is a reading operation and the threshold corresponding to the reading operation is a second threshold count_r, a first number count(read) of media contents having a reading number greater than the second threshold count_r is determined from the R media contents.

[0164] Exemplarily, the electronic device may determine a first number of media contents having a reading volume greater than a second threshold from the R media contents based on the following formula (2):

[0165] count(read) = sum(read_i > count_r) (2)

[0166] Among them, count(read) represents the number of media contents among the R media contents whose reading volume is greater than the second threshold, that is, the first number, count_r is the second threshold corresponding to the reading operation, and based on read_i>count_r, it indicates whether the reading volume of the i-th media content among the R media contents is greater than the second threshold. In the embodiment of the present application, if the reading volume of the media content is greater than the first threshold, the judgment result of read_i>count_r is 1, otherwise the judgment result of read_i>count_r is 0.

[0167] In one example, if the kth object operation is a collection operation and the threshold corresponding to the collection operation is a third threshold count_s, a first number count(save) of media contents whose collection number is greater than the third threshold count_s is determined from the R media contents.

[0168] Exemplarily, the electronic device may determine a first number of media contents whose collection amount is greater than a third threshold from the R media contents based on the following formula (3):

[0169] count(save) = sum(save_i > count_s) (3)

[0170] Among them, count(save) represents the number of media contents among the R media contents whose collection quantity is greater than the third threshold, that is, the first quantity, count_s is the second threshold corresponding to the collection operation, and save_i>count_s represents whether the collection quantity of the i-th media content among the R media contents is greater than the third threshold. In this embodiment of the present application, if the collection quantity of the media content is greater than the third threshold, the judgment result of save_i>count_s is 1, otherwise the judgment result of save_i>count_s is 0.

[0171] Based on the above steps, the electronic device determines a first number of media contents from the R media contents for which operation data for the k-th type of object operation is greater than a corresponding threshold. Next, based on the first number of media contents corresponding to the k-th type of object operation and the number of media contents in R, the electronic device determines a weight corresponding to the k-th type of object operation.

[0172] The embodiment of the present application does not limit the specific manner in which the electronic device determines the weight corresponding to the k-th object operation based on the first number corresponding to the k-th object operation and the number of R media contents.

[0173] In some embodiments, a ratio of the first number corresponding to the k-th object operation to the number of R media contents is determined as the weight corresponding to the k-th object operation.

[0174] In one example, if the k-th object operation is a forwarding operation, the ratio of the first value count(forward) corresponding to the forwarding operation to R is determined as the weight corresponding to the forwarding operation.

[0175] For example, the electronic device may determine the weight corresponding to the forwarding operation according to the following formula (4):

[0176] Wf= count(forward) / R (4)

[0177] Wherein, Wf is the weight corresponding to the forwarding operation, count(forward) is the first value corresponding to the forwarding operation, and R is the number of R media contents.

[0178] In an example, if the k-th object operation is a read operation, the ratio of the first value count(read) corresponding to the read operation to R is determined as the weight corresponding to the read operation.

[0179] For example, the electronic device may determine the weight corresponding to the forwarding operation according to the following formula (5):

[0180] Wr= count(read) / R (5)

[0181] Wherein, Wr is the weight corresponding to the reading operation, count(read) is the first value corresponding to the reading operation, and R is the number of R media contents.

[0182] In one example, if the k-th object operation is a save operation, the ratio of the first value count(save) corresponding to the save operation to R is determined as the weight corresponding to the save operation.

[0183] For example, the electronic device may determine the weight corresponding to the collection operation according to the following formula (6):

[0184] Ws= count(save) / R (6)

[0185] Wherein, Ws is the weight corresponding to the collection operation, count(save) is the first value corresponding to the collection operation, and R is the number of R media contents.

[0186] In an embodiment of the present application, in addition to determining the weight corresponding to the k-th object operation by using the ratio of the first number corresponding to the k-th object operation and the number of R media contents, the electronic device may also use other methods to determine the weight corresponding to the k-th object operation based on the first number corresponding to the k-th object operation and the number of R media contents. For example, the electronic device first determines the ratio of the first number corresponding to the k-th object operation and the number of R media contents, and determines the weight corresponding to the k-th object operation by multiplying the ratio by a preset value, where the preset value is a positive number.

[0187] The above describes the specific process of determining the weight corresponding to the kth object operation among the Q object operations. The above method can be used to determine the weights corresponding to other object operations among the Q object operations.

[0188] After the electronic device determines the weight corresponding to each of the Q object operations based on the above steps, it executes the above step S103-C to determine the popularity value of the i-th media content based on the weights corresponding to the Q object operations and the operation data of the Q object operations of the i-th media content.

[0189] The embodiment of the present application does not limit the specific manner in which the electronic device determines the popularity value of the i-th media content based on the weights corresponding to the Q object operations and the operation data of the Q object operations of the i-th media content.

[0190] In some embodiments, based on the weights corresponding to the Q object operations, at least one object operation with a larger weight is selected from the Q object operations, and then the popularity value of the i-th media content is determined based on the operation data of the at least one object operation among the Q object operations for the i-th media content. For example, the sum of the operation data of the at least one object operation for the i-th media content is determined as the popularity value of the i-th media content.

[0191] In some embodiments, the above S103-C includes the following steps S103-C1 to S103-C3:

[0192] S103-C1, determining a second value corresponding to the kth object operation of the i-th media content based on the operation data of the kth object operation of the i-th media content and a threshold value corresponding to the kth object operation;

[0193] S103-C2, determining a popularity value of the kth object operation of the i-th media content based on the second value corresponding to the kth object operation of the i-th media content and the weight corresponding to the kth object operation;

[0194] S103 - C3 : Determine the popularity value of the i-th media content based on the popularity values ​​of the Q types of object operations on the i-th media content.

[0195] In this implementation, the electronic device can determine the popularity value of each of the Q object operations for the i-th media content based on the weights of the Q object operations and the operation data of the Q object operations for the i-th media content. Furthermore, based on the popularity values ​​of the Q object operations for the i-th media content, the popularity value of the i-th media content can be determined. The specific method for determining the popularity value of each of the Q object operations is essentially the same, and will be described here using the k-th object operation as an example.

[0196] For the kth object operation, the electronic device determines a popularity value of the kth object operation of the ith media content based on the second value corresponding to the kth object operation of the ith media content and the weight corresponding to the kth object operation.

[0197] For example, if the operation data of the k-th object operation of the i-th media content is greater than or equal to the threshold corresponding to the k-th object operation, the second value is determined to be the first preset value.

[0198] For another example, if the operation data of the kth object operation of the i-th media content is less than the threshold corresponding to the kth object operation, the second value is determined to be a second preset value, and the second preset value is less than the first preset value.

[0199] The embodiment of the present application does not limit the specific values ​​of the first preset value and the second preset value, as long as the first preset value is greater than the second preset value.

[0200] In an example, the first preset value is 1, and the second preset value is 0.

[0201] For example, assuming that the k-th object operation is a forwarding operation, and the threshold corresponding to the forwarding operation is threshold 1, the forwarding volume of the i-th media content is obtained. If the forwarding volume of the i-th media content is greater than the threshold 1, the first preset value (for example, 1, of course, it can also be other values) is determined as the second value corresponding to the forwarding operation; if the forwarding volume of the i-th media content is less than or equal to the threshold 1, the second preset value (for example, 0, of course, it can also be other values) is determined as the second value corresponding to the forwarding operation. Alternatively, if the forwarding volume of the i-th media content is greater than or equal to the threshold 1, the first preset value (for example, 1, of course, it can also be other values) is determined as the second value corresponding to the forwarding operation; if the forwarding volume of the i-th media content is less than the threshold 1, the second preset value (for example, 0, of course, it can also be other values) is determined as the second value corresponding to the forwarding operation.

[0202] Assuming that the k-th object operation is a reading operation, and the threshold corresponding to the reading operation is threshold 2, the reading volume of the i-th media content is obtained. If the reading volume of the i-th media content is greater than the threshold 2, the first preset value (for example, 1, of course, it can also be other values) is determined as the second value corresponding to the reading operation. If the reading volume of the i-th media content is less than or equal to the threshold 2, the second preset value (for example, 0, of course, it can also be other values) is determined as the second value corresponding to the reading operation. Alternatively, if the reading volume of the i-th media content is greater than or equal to the threshold 2, the first preset value (for example, 1, of course, it can also be other values) is determined as the second value corresponding to the reading operation; if the reading volume of the i-th media content is less than the threshold 2, the second preset value (for example, 0, of course, it can also be other values) is determined as the second value corresponding to the reading operation.

[0203] The present embodiment does not limit the specific values ​​of the thresholds corresponding to the various object operations. In one example, the thresholds corresponding to the object operations in S103-C1 are the same as the thresholds corresponding to the object operations in S103-B21. For example, the threshold corresponding to the forwarding operation is the first threshold, the threshold corresponding to the reading operation is the second threshold, and the threshold corresponding to the favorite operation is the third threshold.

[0204] In an embodiment of the present application, the second numerical value corresponding to the kth object operation of the ith media content can reflect the popularity of the kth object operation in the ith media content. Based on this, in addition to using the above-mentioned method to determine the second numerical value corresponding to the kth object operation of the ith media content, the electronic device can also compare the operation data of the kth object operation of the ith media content with the corresponding threshold value. If the operation data of the kth object operation of the ith media content is greater than or equal to the threshold value corresponding to the kth object operation, the operation data of the kth object operation of the ith media content is determined as the second numerical value corresponding to the kth object operation of the ith media content. Of course, other methods can also be used to determine the second numerical value corresponding to the kth object operation of the ith media content, and the embodiment of the present application does not limit this.

[0205] Based on the above steps, after determining the second numerical value corresponding to the kth object operation of the i-th media content, execute the above steps S103-C2 to determine the heat value of the kth object operation of the i-th media content based on the second numerical value corresponding to the kth object operation of the i-th media content and the weight corresponding to the kth object operation.

[0206] For example, the product of the second value corresponding to the k-th object operation of the i-th media content and the weight corresponding to the k-th object operation is determined as the heat value of the k-th object operation of the i-th media content.

[0207] For another example, the product of the second numerical value corresponding to the kth object operation of the i-th media content and the weight corresponding to the kth object operation is determined, and then the product of the product and the preset value is determined as the heat value of the kth object operation of the i-th media content, where the preset value is a positive number.

[0208] Based on the above steps, the electronic device can determine the heat value of each object operation among the Q object operations of the i-th media content, and then execute the above steps S103-C3 to determine the heat value of the i-th media content based on the heat values ​​of the Q object operations of the i-th media content.

[0209] The embodiment of the present application does not limit the specific manner in which the electronic device determines the popularity value of the i-th media content based on the popularity values ​​of the Q types of object operations on the i-th media content.

[0210] In a possible implementation, the maximum value of the heat values ​​of the Q types of object operations of the i-th media content is determined as the heat value of the i-th media content.

[0211] In a possible implementation, the sum of the heat values ​​of Q types of object operations on the i-th media content is determined as the heat value of the i-th media content.

[0212] In one example, assuming that the Q types of object operations include forwarding operations, reading operations, and favorite operations, the electronic device can determine the popularity value of the i-th media content according to the following formula (7):

[0213] Hot_i= w_f*(forward_i > count_f) + w_r*(read_i > count_r)+w_s*(save_i > count_s) (7)

[0214] Among them, Hot_i is the heat value of the i-th media content, w_f, w_r and w_s are the weights corresponding to the forwarding operation, reading operation and collection operation respectively, and count_f, count_r and count_s are the thresholds corresponding to the forwarding operation, reading operation and collection operation respectively. forward_i, read_i and save_i are the forwarding volume, reading volume and collection volume of the i-th media content respectively. The () function is a judgment function. If the judgment condition is met, the corresponding second value is 1, otherwise the corresponding second value is 0. For example, if forward_i>count_f, the second value corresponding to the forwarding operation of the i-th media content is equal to 1. If forward_i≤count_f, the second value corresponding to the forwarding operation of the i-th media content is equal to 0. If read_i>count_r, the second value corresponding to the reading operation of the i-th media content is equal to 1. If read_i≤count_r, the second value corresponding to the reading operation of the i-th media content is equal to 0. If save_i>count_s, the second value corresponding to the collection operation of the i-th media content is equal to 1; if save_i≤count_s, the second value corresponding to the collection operation of the i-th media content is equal to 0.

[0215] The above describes a specific process of determining the popularity value of the i-th media content among the N media contents of the first media account. Referring to the above method, the electronic device can determine the popularity value of each media content among the N media contents.

[0216] S104. Based on the popularity value of each media content in the N media content and the tag data under K different tags, determine the tag data of the first media account under K different tags, and search for the target media account based on the tag data of the first media account under K different tags.

[0217] In an embodiment of the present application, when searching for a target media account, the electronic device first determines any candidate media account of the target media account based on the above steps, such as the tag data of each media content in the N media content of the first media account under K different tags, and determines the popularity value of each media content in the N media content. Then, based on the tag data of each media content in the N media content of the first media account under K different tags and the popularity value of each media content in the N media content, the tag data of the first media account under the K tags is determined, and the tag data of the first media account under the K tags is used as a supplementary condition for account search to complete the search of the target media account, which can improve the search accuracy of the media account.

[0218] In other words, the embodiment of the present application extracts and aggregates information from the media content published by the first media account to obtain semantic tags in the media content dimension (i.e., tag data of the media content). Aggregation and screening are then performed based on the tag data of the media content to obtain semantic tags in the account dimension. This can effectively help meet the needs of media account search, expand tags, enhance media account category recall capabilities, and improve media account search accuracy.

[0219] In addition, the present application proposes a method for calculating the popularity of media content based on object operation behavior information such as forwarding operations, favorite operations, and reading operations on the media content, which can accurately calculate the popularity of the media content. In this way, when determining the tag data of the first media account based on the tag data of the media content of the first media account and the popularity value of the media content, the accuracy of determining the tag data of the first media account can be improved.

[0220] The following describes a specific process of determining tag data of a first media account under K different tags based on the popularity value of each media content in N media content and the tag data of each media content under K different tags.

[0221] The embodiment of the present application does not limit the specific manner in which the electronic device determines the tag data of the first media account under K different tags based on the popularity value of each media content in N media content and the tag data of each media content under K different tags.

[0222] In some embodiments, the electronic device may select at least one media content with the highest popularity value from the N media content based on the popularity value of each media content in the N media content, and then determine the tag data of the first media account under K different tags based on the tag data of the at least one media content under K different tags. For example, N1 media content with the highest popularity value may be selected from the N media content, and then the tag data of the first media account under K different tags may be determined based on the tag data of the N1 media content under K different tags.

[0223] In some embodiments, determining the tag data of the first media account under K different tags based on the popularity value of each media content in the N media content and the tag data under K different tags in S104 includes the following steps S104-A and S104-B:

[0224] S104-A: Determine a score for the first media account in each of the P subcategories under the j-th label based on the probability value of each of the P subcategories of the N media content and the popularity value of the N media content;

[0225] S104-B: Determine tag data of the first media account under K different tags based on the scores of the first media account in each subcategory under K different tags.

[0226] In this implementation, the tag data for each piece of N media content under K different tags includes the probability value of each piece of media content belonging to each sub-component under each of the K tags. For example, for the i-th piece of media content among the N pieces of media content and the j-th tag among the K tags, the tag data for the i-th piece of media content under the j-th tag includes the probability value of the i-th piece of media content belonging to each of the P subcategories under the j-th tag. In this case, the score of the first media account in each of the P subcategories under the j-th tag can be determined based on the probability value of each of the P subcategories under the j-th tag for the N pieces of media content and the popularity value of the N pieces of media content.

[0227] The embodiment of the present application does not limit the specific method in which the electronic device determines the score of the first media account in each of the P subcategories under the j-th category label based on the probability value of each of the P subcategories under the j-th category label of N media content and the popularity value of the N media content.

[0228] In one possible implementation, the electronic device selects one media content 1 from the N media contents based on the popularity values ​​of the N media contents, for example, determining the media content with the largest popularity value among the N media contents as media content 1. Then, based on the probability value of each of the P subcategories under the j-th category label of the media content 1, the score of the first media account in each of the P subcategories under the j-th category label is determined. For example, the probability value of each of the P subcategories under the j-th category label of the media content 1 can be directly determined as the score of the first media account in each of the P subcategories under the j-th category label. For another example, the preset value corresponding to the probability value of each of the P subcategories under the j-th category label of the media content 1 can be directly determined as the score of the first media account in each of the P subcategories under the j-th category label, wherein the larger the probability value, the larger the corresponding preset value.

[0229] In a possible implementation, the above S104-A includes the following step S104-A1:

[0230] S104-A1. For the nth subcategory among P subcategories, based on the probability values ​​of N media contents belonging to the nth subcategory respectively and the popularity values ​​of the N media contents, determine the score of the first media account in the nth subcategory, where n is a positive integer less than or equal to P.

[0231] Correspondingly, the above S104-B includes the following steps of S104-B1:

[0232] S104 - B1 . Determine tag data of the first media account under the jth tag category based on the scores of the first media account in the P subcategories of the jth tag category.

[0233] The specific implementation process of the above S104-A1 is first introduced below.

[0234] In the embodiments of the present application, the specific method for determining the score of the first media account in each of the P subcategories under the j-th category label is substantially the same. For ease of description, the n-th subcategories among the P subcategories are used as an example. For the n-th subcategories among the P subcategories, the electronic device can determine the score of the first media account in the n-th subcategories based on the probability value of each piece of N media content belonging to the n-th subcategories and the popularity value of the N media content.

[0235] In one implementation of S104-A1, the score of the first media account in the nth subcategory may be determined by multiplying the probability value of each of the N media contents belonging to the nth subcategory by the popularity value of the N media contents.

[0236] In another implementation of S104-A1, the above S104-A1 may include the following steps S104-A11 and S104-A12:

[0237] S104-A11. Based on the probability values ​​of the N media contents respectively belonging to the nth subcategory, select T media contents from the N media contents, where T is a positive integer less than or equal to N;

[0238] S104-A12: Determine a score of the first media account in the nth subcategory based on the probability values ​​of the T media contents respectively belonging to the nth subcategory and the popularity values ​​of the T media contents.

[0239] In this implementation, to improve the accuracy of determining the score of the first media account in the nth subcategory, the electronic device screens N media contents. Specifically, the electronic device first selects T media contents from the N media contents based on the probability values ​​of the N media contents belonging to the nth subcategory. For example, based on the probability values ​​of the N media contents belonging to the nth subcategory, the electronic device selects T media contents from the N media contents whose probability values ​​of the nth subcategory are greater than or equal to the corresponding preset values. Then, based on the probability values ​​of these T media contents belonging to the nth subcategory and the popularity values ​​of the T media contents, the score of the first media account in the nth subcategory is determined.

[0240] The embodiment of the present application does not limit the specific manner in which the electronic device determines the score of the first media account in the nth subcategory based on the probability values ​​of the T media contents respectively belonging to the nth subcategory and the popularity values ​​of the T media contents.

[0241] In one possible implementation, for each of the T media contents, the probability value that the media content belongs to the nth subcategory is multiplied by the popularity value of the media content to obtain a third value; the third values ​​corresponding to each of the T media contents are added to obtain the score of the first media account in the nth subcategory.

[0242] For example, assuming that the jth tag is a category tag and the nth subcategory is the nth subcategory cate_n in the category tag, the score C_n of the first media account in the nth subcategory cate_n under the category tag can be determined according to the following formula (7):

[0243] C_n = sum(s_j(cate_n)*hot_j) (7)

[0244] Among them, s_j(cate_n) is the probability value that the j-th media content among T media content belongs to the n-th subcategory cate_n under the category signature, hot_j is the heat value of the j-th media content, and s_j(cate_n)*hot_j is the third value corresponding to the j-th media content.

[0245] For example, assume that T media contents include 3 media contents, media content 1, media content 2 and media content 3, the heat values ​​of media content 1, media content 2 and media content 3 are hot1, hot2 and hot3 respectively, the probability that media content 1 belongs to the nth subcategory cate_n is s1, the probability that media content 2 belongs to the nth subcategory cate_n is s2, and the probability that media content 3 belongs to the nth subcategory cate_n is s3. In this way, the probability s1 that media content 1 belongs to the nth cate_n subcategory is multiplied by the heat value hot1 of media content 1 to obtain the third value corresponding to media content 1, the probability s2 that media content 2 belongs to the nth cate_n subcategory is multiplied by the heat value hot2 of media content 2 to obtain the third value corresponding to media content 2, the probability s3 that media content 3 belongs to the nth cate_n subcategory is multiplied by the heat value hot3 of media content 3 to obtain the third value corresponding to media content 3, and then the third values ​​corresponding to media content 1, media content 2 and media content 3 are added respectively to obtain the score of the first media account in the nth subcategory under the category target signature.

[0246] Based on the above steps, the electronic device may determine the score of the first media account in each subcategory under the category signature.

[0247] For example, assuming that the jth tag is an entity tag and the nth subcategory is the nth subcategory entity_n in the entity tag, the score E_n of the first media account in the nth subcategory entity_n under the entity tag can be determined according to the following formula (8):

[0248] E_n = sum(s_j(entity_n)*hot_j) (8)

[0249] Among them, s_j(entity_n) is the probability value of the j-th media content among T media content belonging to the n-th subcategory entity_n signed by the entity, hot_j is the heat value of the j-th media content, and s_j(entity_n)*hot_j is the third value corresponding to the j-th media content.

[0250] For example, assume that T media contents include 3 media contents, media content 1, media content 2 and media content 3, the heat values ​​of media content 1, media content 2 and media content 3 are hot1, hot2 and hot3 respectively, the probability that media content 1 belongs to the nth subcategory entity_n is s11, the probability that media content 2 belongs to the nth subcategory entity_n is s12, and the probability that media content 3 belongs to the nth subcategory entity_n is s13. In this way, the probability s11 that media content 1 belongs to the nth entity_n subcategory is multiplied by the heat value hot1 of media content 1 to obtain the third value corresponding to media content 1, the probability s12 that media content 2 belongs to the nth entity_n subcategory is multiplied by the heat value hot2 of media content 2 to obtain the third value corresponding to media content 2, and the probability s13 that media content 3 belongs to the nth entity_n subcategory is multiplied by the heat value hot3 of media content 3 to obtain the third value corresponding to media content 3. Then, the third values ​​corresponding to media content 1, media content 2 and media content 3 are added together to obtain the score of the first media account in the nth subcategory under the entity label.

[0251] Based on the above steps, the electronic device can determine the score of the first media account in each subcategory under the entity tag.

[0252] In an embodiment of the present application, the electronic device determines the scores of the first media account in P subcategories under the jth category based on the above method, and then determines the tag data of the first media account under the jth category label based on the scores of the first media account in each subcategories under the jth category label.

[0253] The embodiment of the present application does not limit the specific manner in which the electronic device determines the tag data of the first media account under the j-th category tag based on the score of the first media account in each subcategory under the j-th category tag.

[0254] In some embodiments, the electronic device may select at least one subcategory from the P subcategories under the j-th category of the first media account based on the first media account's scores in the P subcategories under the j-th category of the label, and determine the selected subcategory as the label data for the first media account under the j-th category of the label. For example, based on the first media account's scores in the P subcategories under the j-th category of the label, the electronic device may select the subcategory with the highest score from the P subcategories and determine the selected subcategory as the label data for the first media account under the j-th category of the label.

[0255] In some embodiments, the above S104-B1 includes the following steps S104-B11 to S104-B13:

[0256] S104-B11, determining the sum of the scores of the first media account in P subcategories under the j-th label;

[0257] S104-B12: For the nth subcategory, determine a proportion of the nth subcategory based on a ratio of the score of the first media account in the nth subcategory to the sum of the scores;

[0258] S104-B13: Based on the proportion of the first media account in each of the P subcategories, select at least one subcategory from the P subcategories and determine it as the label data of the first media account under the j-th label.

[0259] In this implementation, when determining the label data of the i-th media content under the j-th category label, first determine the proportion of each subcategory under the j-th category label, and then based on the proportion of each subcategory and the score of the first media account in each subcategory, determine the label data of the first media account under the j-th category label.

[0260] Specifically, based on the first media account's scores in P subcategories under the j-th category, the sum of the first media account's scores in the P subcategories under the j-th category is determined. For the n-th subcategory among the P subcategories, the proportion of the n-th subcategory is determined based on the ratio of the first media account's score in the n-th subcategory to the sum of the scores.

[0261] In one example, assuming that the j-th tag is a category tag, the sum of the scores of the first media account in each subcategory under the category tag is determined by the following formula (9):

[0262] Score_c = sum(C_n) (9)

[0263] Among them, Score_c is the sum of the scores of the first media account in each subcategory signed by the target of this type, and C_n is the score of the first media account in the nth subcategory signed by the target of this type.

[0264] For example, the electronic device may determine the proportion of the nth subcategory under the target category signature by the following formula (10):

[0265] Rate_c_n = C_n / score_c (10)

[0266] Among them, Rate_c_n is the proportion of the nth subcategory signed by the class target.

[0267] Based on the above steps, the electronic device can determine the proportion of each sub-category under the category signature.

[0268] In one example, assuming that the j-th category label is an entity label, the sum of the scores of the first media account in each subcategory under the entity label is determined by the following formula (11):

[0269] Score_e = sum(E_n) (11)

[0270] Among them, Score_e is the sum of the scores of the first media account in each subcategory under the entity tag, and E_n is the score of the first media account in the nth subcategory under the entity tag.

[0271] For example, the electronic device may determine the proportion of the nth subcategory under the entity tag by using the following formula (12):

[0272] Rate_e_n = e_n / score_e (12)

[0273] Among them, Rate_e_n is the proportion of the nth subcategory under the entity label.

[0274] Based on the above steps, the electronic device can determine the proportion of each sub-category under the category signature.

[0275] In this way, the electronic device can calculate the proportion of each subcategory in the P subcategories, and then based on the proportion of the first media account in each subcategory in the P subcategories, select at least one subcategory from the P subcategories and determine it as the label data of the first media account under the jth category label.

[0276] In one possible implementation, based on the proportion of the first media account in each of the P subcategories, at least one subcategories whose proportion is greater than a preset proportion is selected from the P subcategories and determined as the label data of the first media account under the jth category label.

[0277] In one possible implementation, based on the score of the first media account in each of the P subcategories and the proportion of each subcategory, at least one subcategory is selected from the P subcategories and determined as the label data of the first media account under the jth category label.

[0278] For example, based on the score of the first media account in each of the P subcategories and the proportion of each subcategory, at least one subcategory with a score greater than a preset score value and a proportion greater than a preset proportion value is selected from the P subcategories and determined as the label data of the first media account under the jth category label.

[0279] In one example, assuming that the jth tag is a category tag, the tag data of the first media account under the category tag can be determined by the following formula (13):

[0280] C_n>=cate_x

[0281] Rate_c_n >= rate_c_x (12)

[0282] Where cate_x is the preset score value corresponding to the category tag, and rate_c_x is the preset percentage value corresponding to the category tag. Based on the scores of each subcategory under the category tag of the first media account and the percentage values ​​of each subcategory under the category tag, at least one subcategory that satisfies the above formula (12) can be selected from each subcategory under the category tag and determined as the tag data of the first media account under the category tag.

[0283] In some embodiments, if for the two conditions shown in the above formula (12), if the scores of the subcategories of the first media account under the category target signature do not satisfy the condition C_n>=cate_x (that is, none of the scores of the subcategories of the first media account under the category target signature is greater than the preset score value), then the at least one subcategory can be selected from the P subcategories based on the condition Rate_c_n>=rate_c_x, and determined as the label data of the first media account under the category target signature. On the contrary, if the proportion values ​​of the subcategories under the category target signature do not satisfy the condition Rate_c_n>=rate_c_x (that is, none of the proportion values ​​of the subcategories under the category target signature is greater than the preset proportion value), then the at least one subcategory can be selected from the P subcategories based on the condition C_n>=cate_x, and determined as the label data of the first media account under the category target signature.

[0284] In one example, assuming that the jth type of tag is an entity tag, the tag data of the first media account under the entity tag can be determined by the following formula (13):

[0285] E_n>=E_x

[0286] Rate_e_n >= rate_e_x (13)

[0287] Where cate_x is the preset score value corresponding to the entity tag, and rate_e_x is the preset percentage value corresponding to the entity tag. Based on the scores of each subcategory under the entity tag of the first media account and the percentage values ​​of each subcategory under the entity tag, at least one subcategory that satisfies the above formula (13) can be selected from each subcategory under the entity tag and determined as the tag data of the first media account under the entity tag.

[0288] In some embodiments, if for the two conditions shown in formula (13) above, if the scores of the subcategories of the first media account under the entity tag do not satisfy the condition E_n>=E_x (i.e., none of the scores of the subcategories of the first media account under the entity tag is greater than a preset score value), then the at least one subcategories can be selected from the P subcategories based on the condition Rate_e_n>=rate_e_x and determined as the tag data of the first media account under the category tag. On the contrary, if the proportion values ​​of the subcategories under the entity tag do not satisfy the condition Rate_e_n>=rate_e_x (i.e., none of the proportion values ​​of the subcategories under the entity tag is greater than a preset proportion value), then the at least one subcategories can be selected from the P subcategories based on the condition E_n>=E_x and determined as the tag data of the first media account under the category tag.

[0289] In this embodiment of the present application, the tag data for each of the K different tag categories of the first media account can be determined based on the above steps. Thus, when subsequently recalling target media accounts, the number of tags for each of the K different tag categories of the first media account can be used as the account dimension information of the first media account to expand the tags of the first media account, thereby improving the search accuracy and search capability of the media account.

[0290] For example, for each of the multiple candidate media accounts corresponding to the target media account, the method of the embodiment of the present application can be used to determine the label data of each media account in these multiple candidate media accounts, and then search for the target media account based on the label data of these multiple candidate media accounts, which can improve the search accuracy of the target media account.

[0291] In some embodiments, in the embodiments of the present application, if the above electronic device is Figure 1When the server in the target media account is retrieved, the search results of the target media account returned by the server to the terminal device may include at least one media content published by the target media account. In a possible implementation, the at least one media content of the target media account recalled above may be obtained by sorting and recalling based on the popularity value of the media content published by the target media account. Specifically, the electronic device obtains multiple media contents of the target media account, for example, obtains multiple media contents published by the target media account within a preset time period, and / or recalls multiple media contents based on the relevant search terms in the search request. The operation data of Q types of object operations of each media content in the multiple media contents are obtained, and the popularity value of the media content is determined for the operation data of Q types of object operations of each media content in the multiple media contents. The specific process can refer to the description of the above-mentioned heat value determination process and will not be repeated here. In this way, the electronic device can sort the multiple media contents based on the popularity value of each media content in the multiple media contents of the target media account, for example, arranging the media content with a high popularity value in front, and sending the sorted media content to the terminal device for display.

[0292] The media account search method provided in the embodiment of the present application, when searching for a target media account, obtains N media content published by any candidate media account corresponding to the target media account, such as the first media account. For the i-th media content in the N media content, based on the content information of the i-th media content, the tag data of the i-th media content under K different tags is determined. At the same time, based on the operation data of Q types of object operations on the i-th media content, the popularity value of the i-th media content is determined. Finally, based on the popularity value of each media content in the N media content and the tag data of each media content under K different tags, the tag data of the first media account under K different tags is determined, and then the target media account search is performed based on the tag data of the first media account under K different tags. In other words, the embodiment of the present application determines the tag data of the media content dimension based on the media content of the first media account, and then determines the tag data of the account dimension based on the popularity value of the media content and the tag data of the media content dimension, so as to expand the tags of the first media account, thereby improving the search accuracy and search capability of the media account. In addition, the present application also proposes a method for calculating the popularity value of media content based on Q types of object operations on media content, which can accurately calculate the popularity value of media content. In this way, when determining the tag data of the first media account based on the tag data of the media content of the first media account and the popularity value of the media content, the accuracy of determining the tag data of the first media account can be improved. When searching for media accounts based on the accurately determined tag data, the search accuracy of the media account can be further improved.

[0293] The above is an overall introduction to the search process of media accounts in the embodiment of the present application. Figure 5 ,The search process of media accounts when K class labels are used as category labels and entity labels is further introduced.

[0294] Figure 5 This is a flowchart of a method for searching media accounts provided by an embodiment of the present application. In this embodiment, the method for searching media accounts provided by an embodiment of the present application is introduced by taking K-category tags including category tags and entity tags as an example.

[0295] like Figure 5 As shown, the method of the embodiment of the present application includes the following steps:

[0296] S201 : Based on a search request of a target media account, obtain N media contents published by a first media account.

[0297] The first media account is any candidate media account of the target media account, and N is a positive integer.

[0298] The specific implementation process of the above S201 can refer to the relevant description of the above S101, which will not be repeated here.

[0299] S202 : For an i-th media content among N media content, determine tag data of the i-th media content under a category tag and an entity tag based on content information of the i-th media content.

[0300] Wherein, i is a positive integer less than or equal to N, and K is a positive integer.

[0301] For example, based on the content information of the i-th media content, the probability value of the i-th media content belonging to each of the subcategories under the category target is determined. Then, the probability value of the i-th media content belonging to each of the subcategories under the category target is determined as the label data of the i-th media content under the category target. For example, Figure 3 As shown, the content information of the i-th media content is classified by the classification model corresponding to the category label to obtain the probability value of the i-th media content belonging to each subcategory of the subcategories under the category label.

[0302] For another example, based on the content information of the i-th media content, the probability value of the i-th media content belonging to each of the subcategories under the category tag is determined. Then, the probability value of the i-th media content belonging to each of the subcategories under the entity tag is determined as the label data of the i-th media content under the entity tag. For example, Figure 4As shown, the content information of the i-th media content is classified by the classification model corresponding to the entity tag to obtain the probability value of the i-th media content belonging to each subcategory of the subcategories under the entity tag.

[0303] The specific implementation process of the above S202 can refer to the relevant description of the above S102 and will not be repeated here.

[0304] S203 : Determine the popularity value of the i-th media content based on the operation data of the Q types of object operations on the i-th media content.

[0305] Wherein, Q is a positive integer.

[0306] The Q types of object operations include at least one of forwarding, reading, and collecting.

[0307] In some embodiments, the popularity value of the i-th media content can be determined based on the weights of the Q types of object operations and the operation data of the Q types of object operations on the i-th media content. Specifically, M media content published by a first media account is obtained, where M is a positive integer greater than 1; based on the operation data of the Q types of object operations on the M media content, the weights corresponding to the Q types of object operations are determined; and based on the weights corresponding to the Q types of object operations and the operation data of the Q types of object operations on the i-th media content, the popularity value of the i-th media content is determined.

[0308] The specific implementation process of the above S203 can refer to the relevant description of the above S103, which will not be repeated here.

[0309] S204: Determine the tag data under the category tag and the tag data under the entity tag of the first media account based on the popularity value of each media content in the N media content and the tag data under the category tag and the tag data under the entity tag of each media content.

[0310] For example, based on the probability value of each subcategory of N media contents in each subcategory under the category target signature, and the popularity value of N media contents, the score of the first media account in each subcategory under the category target signature is determined; and then based on the score of the first media account in each subcategory under the category target signature, the label data of the first media account under the category target signature is determined. In some embodiments, for the nth subcategory under the category target signature, the electronic device first selects T media contents from the N media contents based on the probability value of the N media contents belonging to the nth subcategory under the category label, and determines the score of the first media account in the nth subcategory based on the probability value of the T media contents belonging to the nth subcategory and the popularity value of the T media contents. For example, for each of the T media contents, the probability value of the media content belonging to the nth subcategory under the category target signature is multiplied by the popularity value of the media content to obtain a third value. Then, the third values ​​corresponding to each of the T media contents are added together to obtain the score of the first media account in the nth subcategory under the category target signature.

[0311] By referring to the above method, the scores of the first media account in each subcategory under the category target signature can be determined, and then based on the scores of the first media account in each subcategory under the category target signature, the tag data of the first media account under the category target signature can be determined. For example, the electronic device first determines the sum of the scores of the first media account in each subcategory under the category target signature; for the nth subcategory under the category target signature, based on the ratio of the score of the first media account in the nth subcategory to the sum of the above scores, the proportion of the nth subcategory is determined, and then based on the proportion of the first media account in each of the subcategories under the category target signature, at least one subcategory is selected from the subcategories under the category target signature and determined as the tag data of the first media account under the category target signature.

[0312] For another example, based on the probability value of each subcategory of N media contents under the entity tag and the popularity value of the N media contents, the score of the first media account in each subcategory under the entity tag is determined; and then based on the score of the first media account in each subcategory under the entity tag, the tag data of the first media account under the entity tag is determined. In some embodiments, for the nth subcategory under the entity tag, the electronic device first selects T media contents from the N media contents based on the probability value of the N media contents belonging to the nth subcategory under the category tag, and determines the score of the first media account in the nth subcategory based on the probability value of the T media contents belonging to the nth subcategory and the popularity value of the T media contents. For example, for each of the T media contents, the probability value of the media content belonging to the nth subcategory under the entity tag is multiplied by the popularity value of the media content to obtain a third value. Then, the third values ​​corresponding to each of the T media contents are added to obtain the score of the first media account in the nth subcategory under the entity tag.

[0313] By referring to the above method, the scores of the first media account in each subcategory under the entity tag can be determined, and then, based on the scores of the first media account in each subcategory under the entity tag, the tag data of the first media account under the entity tag can be determined. For example, the electronic device first determines the sum of the scores of the first media account in each subcategory under the entity tag; for the nth subcategory under the entity tag, based on the ratio of the score of the first media account in the nth subcategory to the sum of the above scores, the proportion of the nth subcategory is determined; then, based on the proportion of each subcategory of the first media account under the entity tag, at least one subcategory is selected from the subcategories under the entity tag and determined as the tag data of the first media account under the entity tag.

[0314] S205 : Search for a target media account based on the tag data of the first media account under the category tag and the tag data under the entity tag.

[0315] Based on the above steps, the label data under the category label and the label data under the entity label of each candidate media account of the target media account can be determined, and then the target media account can be searched based on the label data under the category label and the label data under the entity label of these candidate media accounts, as well as some other characteristic information of these media accounts (such as at least one of the information such as the name of the media account, the introduction of the media account, etc.), which can improve the search accuracy of the media account.

[0316] In the embodiment of the present application, when searching for a target media account, for at least one candidate media account in the media account, such as the first media account, information is extracted and aggregated from the media content published by the first media account to obtain semantic labels of the media content dimension (i.e., label data of the media content). Aggregation and screening are then performed based on the label data of the media content to obtain semantic labels of the account dimension, which can effectively help the media account search requirements to expand labels, strengthen the media account recall capability, and improve the search accuracy of the media account. In addition, the embodiment of the present application also proposes a method for calculating the heat value of the media content based on object operation behavior information such as forwarding operations, collection operations, and reading operations of the media content, which can achieve accurate calculation of the heat value of the media content. In this way, when determining the label data of the first media account based on the label data of the media content of the first media account and the heat value of the media content, the accuracy of determining the label data of the first media account can be improved, and when searching for media accounts based on the accurately determined label data, the search accuracy of the media account can be further improved.

[0317] Combined with the above Figures 2 to 5 , describes the method embodiment of the present application in detail, and the following is combined with Figure 6 , describe in detail the device embodiments of the present application.

[0318] Figure 6 FIG1 is a schematic block diagram of a media account search device provided in an embodiment of the present application. The device 10 can be applied to electronic devices.

[0319] like Figure 6 As shown, the media account search device 10 includes:

[0320] An acquiring unit 11 is configured to acquire N media contents published by a first media account based on a search request of a target media account, where the first media account is any candidate media account of the target media account, and N is a positive integer;

[0321] a content label determination unit 12 configured to determine, for an i-th media content among the N media content, label data of the i-th media content under K different labels based on content information of the i-th media content, where i is a positive integer less than or equal to N, and K is a positive integer;

[0322] a popularity determining unit 13, configured to determine a popularity value of the i-th media content based on operation data of Q types of object operations on the i-th media content, where Q is a positive integer;

[0323] The account tag determination unit 14 is used to determine the tag data of the first media account under the K different tags based on the popularity value of each media content in the N media content and the tag data under the K different tags, and perform the target media account search based on the tag data of the first media account under the K different tags.

[0324] In some embodiments, the content label determination unit 12 is specifically configured to determine, for a j-th label among the K different labels, based on the content information of the i-th media content, a probability value that the i-th media content belongs to each of the P subcategories under the j-th label, where j is a positive integer less than or equal to K, and P is a positive integer; and determine label data for the i-th media content under the j-th label based on the probability value that the i-th media content belongs to each of the P subcategories under the j-th label.

[0325] In some embodiments, the content label determination unit 12 is specifically configured to classify the content information of the i-th media content using the classification model corresponding to the j-th label to obtain a probability value of the i-th media content belonging to each of the P subcategories under the j-th label.

[0326] In some embodiments, the content tag determination unit 12 is specifically used to obtain M media contents published by the first media account, where M is a positive integer greater than 1; based on the operation data of Q types of object operations of the M media contents, determine the weights corresponding to the Q types of object operations; based on the weights corresponding to the Q types of object operations and the operation data of the Q types of object operations of the i-th media content, determine the popularity value of the i-th media content.

[0327] In some embodiments, the content tag determination unit 12 is specifically used to select R media data from the M media content based on the number of clicks on each media content in the M media content, where R is a positive integer less than or equal to M; and determine the weights corresponding to the Q object operations based on the operation data of the Q object operations of the R media content.

[0328] In some embodiments, the content label determination unit 12 is specifically used to determine, for a kth object operation among the Q object operations, a first number of media contents whose operation data of the kth object operation is greater than a corresponding threshold from the R media contents, where k is a positive integer less than or equal to Q; and determine a weight corresponding to the kth object operation based on the first number corresponding to the kth object operation and the number of the R media contents.

[0329] In some embodiments, the content label determination unit 12 is specifically configured to determine a ratio of the first number corresponding to the k-th object operation to the number of the R media contents as the weight corresponding to the k-th object operation.

[0330] In some embodiments, the heat determination unit 13 is specifically used to determine the second numerical value corresponding to the kth object operation of the ith media content based on the operation data of the kth object operation of the ith media content and the threshold value corresponding to the kth object operation; determine the heat value of the kth object operation of the ith media content based on the second numerical value corresponding to the kth object operation of the ith media content and the weight corresponding to the kth object operation; determine the heat value of the ith media content based on the heat values ​​of the Qth object operations of the ith media content.

[0331] In some embodiments, the heat determination unit 13 is specifically used to determine that the second value is a first preset value if the operation data of the kth object operation of the i-th media content is greater than or equal to the threshold corresponding to the k-th object operation; if the operation data of the kth object operation of the i-th media content is less than the threshold corresponding to the k-th object operation, then determine that the second value is a second preset value, and the second preset value is less than the first preset value.

[0332] In some embodiments, the heat determination unit 13 is specifically used to determine the product of the second numerical value corresponding to the kth object operation of the i-th media content and the weight corresponding to the kth object operation as the heat value of the kth object operation of the i-th media content.

[0333] In some embodiments, the popularity determination unit 13 is specifically configured to determine the sum of the popularity values ​​of Q types of object operations of the i-th media content as the popularity value of the i-th media content.

[0334] In some embodiments, if the label data of the i-th media content under the j-th label includes the probability value of the i-th media content belonging to each of the P subcategories under the j-th label, the account label determination unit 14 is specifically used to determine the score of the first media account in each of the P subcategories under the j-th label based on the probability value of each of the P subcategories under the j-th label of the N media content and the popularity value of the N media content; and determine the label data of the first media account under the K different labels based on the score of the first media account in each subcategories under the K different labels.

[0335] In some embodiments, the account label determination unit 14 is specifically used to determine the score of the first media account in the nth subcategory among the P subcategories based on the probability values ​​of the N media contents respectively belonging to the nth subcategory and the popularity values ​​of the N media contents, where n is a positive integer less than or equal to P; and determine the label data of the first media account under the jth category label based on the scores of the first media account in the P subcategories in the jth category label.

[0336] In some embodiments, the account label determination unit 14 is specifically used to select T media content from the N media contents based on the probability values ​​that the N media contents respectively belong to the nth subcategory, where T is a positive integer less than or equal to N; and determine the score of the first media account in the nth subcategory based on the probability values ​​that the T media contents respectively belong to the nth subcategory and the popularity values ​​of the T media contents.

[0337] In some embodiments, the account label determination unit 14 is specifically used to multiply the probability value of each of the T media contents that the media content belongs to the nth subcategory by the popularity value of the media content to obtain a third value; and add the third values ​​corresponding to each of the T media contents to obtain the score of the first media account in the nth subcategory.

[0338] In some embodiments, the account label determination unit 14 is specifically used to determine the sum of the scores of the first media account in the P subcategories under the j-th category label; for the n-th subcategories, based on the ratio of the score of the first media account in the n-th subcategories to the sum of the scores, determine the proportion of the n-th subcategories; based on the proportion of the first media account in each of the P subcategories, select at least one subcategories from the P subcategories and determine it as the label data of the first media account under the j-th category label.

[0339] In some embodiments, the account label determination unit 14 is specifically used to select the at least one subcategory from the P subcategories based on the score of the first media account in each of the P subcategories and the proportion of each subcategory, and determine it as the label data of the first media account under the j-th category label.

[0340] It should be understood that the device embodiment and the method embodiment may correspond to each other, and similar descriptions may refer to the method embodiment. To avoid repetition, they will not be described here. Specifically, Figure 6The device shown can execute the embodiment of the above-mentioned media account search method, and the aforementioned and other operations and / or functions of each module in the device are respectively for implementing the corresponding method embodiments of the electronic device. For the sake of brevity, they are not repeated here.

[0341] The apparatus of the embodiment of the present application is described above from the perspective of functional modules in conjunction with the accompanying drawings. It should be understood that the functional module can be implemented in hardware form, can be implemented by instructions in software form, or can be implemented by a combination of hardware and software modules. Specifically, the steps of the method embodiment in the embodiment of the present application can be completed by the hardware integrated logic circuit and / or software form instructions in the processor, and the steps of the method disclosed in the embodiment of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. Optionally, the software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps in the above method embodiment in conjunction with its hardware.

[0342] Figure 7 is a schematic block diagram of an electronic device provided in an embodiment of the present application, Figure 7 The electronic device can be used to execute the above method embodiment.

[0343] like Figure 7 As shown, the electronic device 30 may include:

[0344] The memory 31 and the processor 32 are configured to store a computer program 33 and transmit the program code 33 to the processor 32. In other words, the processor 32 can call and run the computer program 33 from the memory 31 to implement the method in the embodiment of the present application.

[0345] For example, the processor 32 may be configured to execute the steps of the above method according to the instructions in the computer program 33 .

[0346] In some embodiments of the present application, the processor 32 may include but is not limited to:

[0347] General-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.

[0348] In some embodiments of the present application, the memory 31 includes but is not limited to:

[0349] Volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus random access memory (DR RAM).

[0350] In some embodiments of the present application, the computer program 33 may be divided into one or more modules, which are stored in the memory 31 and executed by the processor 32 to implement the method for recording a page provided by the present application. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 33 in the electronic device.

[0351] like Figure 7 As shown, the electronic device 30 may further include:

[0352] The transceiver 34 may be connected to the processor 32 or the memory 31 .

[0353] The processor 32 may control the transceiver 34 to communicate with other devices. Specifically, the processor 32 may send information or data to other devices or receive information or data sent by other devices. The transceiver 34 may include a transmitter and a receiver. The transceiver 34 may further include one or more antennas.

[0354] It should be understood that the various components in the electronic device 30 are connected via a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus and a status signal bus.

[0355] According to one aspect of the present application, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a computer, the computer is enabled to perform the method of the above-mentioned method embodiment. Alternatively, the present application also provides a computer program product containing instructions. When the computer is executed by the instructions, the computer is enabled to perform the method of the above-mentioned method embodiment.

[0356] According to another aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program including 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 executes the computer instructions, causing the computer device to perform the method of the above-described method embodiment.

[0357] In other words, when implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0358] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0359] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0360] Modules described as separate components may or may not be physically separate, and components displayed as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected based on actual needs to achieve the purpose of the present embodiment. For example, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module.

[0361] The above content is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for searching a media account, characterized in that: include: Based on the search request for the target media account, obtain N media content published by a first media account, where the first media account is any candidate media account of the target media account, and N is a positive integer; For an i-th media content among the N media content, determining label data of the i-th media content under K different labels based on content information of the i-th media content, where i is a positive integer less than or equal to N, and K is a positive integer; Determine a popularity value of the i-th media content based on operation data of Q types of object operations on the i-th media content, where Q is a positive integer; Based on the popularity value of each media content in the N media content and the tag data under the K different tags, the tag data of the first media account under the K different tags are determined, and based on the tag data of the first media account under the K different tags, the target media account search is performed.

2. The method according to claim 1, characterized in that The determining, based on the content information of the i-th media content, the label data of the i-th media content under the K different labels includes: For a j-th label among the K different labels, determining, based on the content information of the i-th media content, a probability value that the i-th media content belongs to each of P subcategories under the j-th label, where j is a positive integer less than or equal to K, and P is a positive integer; Based on the probability value that the i-th media content belongs to each of the P subcategories under the j-th category label, label data of the i-th media content under the j-th category label is determined.

3. The method according to claim 2, characterized in that The determining, based on the content information of the i-th media content, a probability value of the i-th media content belonging to each of the P subcategories under the j-th category label includes: The content information of the ith media content is classified using the classification model corresponding to the jth category label to obtain a probability value of the ith media content belonging to each of the P subcategories under the jth category label.

4. The method according to claim 1, wherein The determining the popularity value of the i-th media content based on the operation data of the Q types of object operations on the i-th media content includes: Obtain M media content published by the first media account, where M is a positive integer greater than 1; Determining weights corresponding to the Q types of object operations based on the operation data of the Q types of object operations on the M media contents; The popularity value of the i-th media content is determined based on the weights corresponding to the Q types of object operations and the operation data of the Q types of object operations on the i-th media content.

5. The method according to claim 4, characterized in that The determining weights corresponding to the Q types of object operations based on the operation data of the Q types of object operations on the M media contents includes: Selecting R pieces of media data from the M pieces of media content based on the number of clicks on each piece of media content, where R is a positive integer less than or equal to M; Based on the operation data of the Q types of object operations on the R media contents, weights corresponding to the Q types of object operations are determined.

6. The method according to claim 5, characterized in that The determining, based on the operation data of the Q types of object operations on the R media contents, weights corresponding to the Q types of object operations respectively includes: For a k-th object operation among the Q object operations, determining a first number of media contents having operation data of the k-th object operation greater than a corresponding threshold from the R media contents, where k is a positive integer less than or equal to Q; A weight corresponding to the k-th object operation is determined based on the first number corresponding to the k-th object operation and the number of the R media contents.

7. The method according to claim 6, characterized in that The determining, based on the first number corresponding to the k-th object operation and the number of the R media contents, a weight corresponding to the k-th object operation includes: A ratio of the first number corresponding to the k-th object operation to the number of the R media contents is determined as a weight corresponding to the k-th object operation.

8. The method according to claim 6, characterized in that The determining the popularity value of the i-th media content based on the weights corresponding to the Q types of object operations and the operation data of the Q types of object operations on the i-th media content includes: determining, based on the operation data of the kth object operation of the i-th media content and a threshold value corresponding to the kth object operation, a second value corresponding to the kth object operation of the i-th media content; determining a popularity value of the k-th object operation of the i-th media content based on a second value corresponding to the k-th object operation of the i-th media content and a weight corresponding to the k-th object operation; The popularity value of the i-th media content is determined based on the popularity values ​​of the Q types of object operations on the i-th media content.

9. The method according to claim 8, characterized in that The determining, based on the operation data of the kth object operation of the i-th media content and a threshold value corresponding to the kth object operation, a second value corresponding to the kth object operation of the i-th media content includes: If the operation data of the k-th object operation of the i-th media content is greater than or equal to the threshold corresponding to the k-th object operation, determining that the second value is the first preset value; If the operation data of the kth object operation of the i-th media content is less than the threshold corresponding to the kth object operation, the second value is determined to be a second preset value, and the second preset value is less than the first preset value.

10. The method according to claim 8, characterized in that The determining, based on the second value corresponding to the k-th object operation of the i-th media content and the weight corresponding to the k-th object operation, a popularity value of the k-th object operation of the i-th media content includes: A product of a second value corresponding to a k-th object operation of the i-th media content and a weight corresponding to the k-th object operation is determined as a heat value of the k-th object operation of the i-th media content.

11. The method according to claim 8, characterized in that The determining the popularity value of the i-th media content based on the popularity values ​​of the Q types of object operations on the i-th media content includes: The sum of the heat values ​​of the Q types of object operations on the i-th media content is determined as the heat value of the i-th media content.

12. The method according to any one of claims 2 to 3, characterized in that: If the label data of the i-th media content under the j-th label includes a probability value of the i-th media content belonging to each of P subcategories under the j-th label, determining the label data of the first media account under the K different labels based on the popularity value of each media content in the N media contents and the label data under the K different labels includes: Determining a score for the first media account in each of the P subcategories under the j-th label based on a probability value of each of the P subcategories of the N media contents under the j-th label and a popularity value of the N media contents; Based on the scores of the first media account in each subcategory under the K different tags, tag data of the first media account under the K different tags are determined.

13. The method according to claim 12, characterized in that Determining a score of the first media account in each of the P subcategories under the j-th category label based on the probability value of each of the P subcategories of the N media contents under the j-th category label and the popularity value of the N media contents includes: For an nth subcategory among the P subcategories, determining a score of the first media account in the nth subcategory based on the probability values ​​of the N media contents respectively belonging to the nth subcategory and the popularity values ​​of the N media contents, where n is a positive integer less than or equal to P; The determining, based on the score of the first media account in each subcategory of the K different tags, tag data of the first media account under the K different tags includes: Based on the scores of the first media account in the P subcategories in the j-th category of tags, tag data of the first media account under the j-th category of tags is determined.

14. The method according to claim 13, characterized in that The determining, based on the probability values ​​that the N media contents respectively belong to the nth subcategory and the popularity values ​​of the N media contents, a score of the first media account in the nth subcategory includes: Selecting T media contents from the N media contents based on probability values ​​that the N media contents respectively belong to the n-th subcategory, where T is a positive integer less than or equal to N; Based on the probability values ​​that the T media contents respectively belong to the nth subcategory and the popularity values ​​of the T media contents, a score of the first media account in the nth subcategory is determined.

15. The method according to claim 14, characterized in that Determining the score of the first media account in the nth subcategory based on the probability values ​​that the T media contents respectively belong to the nth subcategory and the popularity values ​​of the T media contents includes: For each piece of media content among the T pieces of media content, multiply the probability value of the media content belonging to the nth subcategory by the popularity value of the media content to obtain a third value; The third values ​​corresponding to each of the T media contents are added together to obtain a score of the first media account in the nth subcategory.

16. The method according to claim 13, characterized in that The determining, based on the scores of the first media account in the P subcategories of the j-th category of tags, the tag data of the first media account under the j-th category of tags includes: Determine a sum of the scores of the first media account in P subcategories under the j-th label; For the nth subcategory, determining a proportion of the nth subcategory based on a ratio of the score of the first media account in the nth subcategory to the sum of the scores; Based on the proportion of the first media account in each of the P subcategories, at least one subcategory is selected from the P subcategories and determined as the label data of the first media account under the j-th label.

17. The method according to claim 16, characterized in that The selecting at least one subcategory from the P subcategories based on the proportion of the first media account in each of the P subcategories and determining it as the tag data of the first media account under the j-th tag includes: Based on the score of the first media account in each of the P subcategories and the proportion of each subcategory, at least one subcategory is selected from the P subcategories and determined as the label data of the first media account under the j-th category label.

18. A device for searching media accounts, characterized in that: include: an acquiring unit, configured to acquire, based on a search request of a target media account, N media contents published by a first media account, where the first media account is any candidate media account of the target media account, and N is a positive integer; a content label determination unit, configured to determine, for an i-th media content among the N media content, label data of the i-th media content under K different labels based on content information of the i-th media content, where i is a positive integer less than or equal to N, and K is a positive integer; a popularity determining unit, configured to determine a popularity value of the i-th media content based on operation data of Q types of object operations on the i-th media content, where Q is a positive integer; An account label determination unit is used to determine the label data of the first media account under the K different labels based on the popularity value of each media content in the N media content and the label data under the K different labels, and to search for the target media account based on the label data of the first media account under the K different labels.

19. A computer device comprising a processor and a memory; The memory is used to store computer programs; The processor is configured to execute the computer program to implement the method according to any one of claims 1 to 17.

20. A computer-readable storage medium, characterized in that For storing computer programs; The computer program enables a computer to execute the method according to any one of claims 1 to 17.

Images