Content search method and device, equipment, medium and program product

Abstract

The invention discloses a content search method and device, equipment, a medium and a program product, and relates to the technical field of computers, the method comprises the following steps: receiving a search request, the search request comprising first search content, and the first search content is a search condition during data search; associated search content corresponding to the first search content is obtained, the associated search content is search content within a historical time period, the search trigger interval duration of the first search content and the associated search content is within a preset duration range, and the first search content and the associated search content meet a preset association condition; and searching the first search content and the associated search content to obtain a search result. According to the method, the result obtained when the associated search content is searched can be continuously added on the basis of the original search content, the more accurate and comprehensive search result is fed back to the user under the condition that the intention of the user is fully understood, and the search accuracy of the user based on the search content is improved.

Description

Technical Field

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

[0002] Search engines are important tools for information retrieval on the Internet. They use complex algorithms to index multimedia content on the Internet, allowing users to quickly locate information of interest through keywords.

[0003] In the related art, search engines usually provide a user input area, in which the user types the keywords to be searched, and the search engine matches the corresponding content from the database according to the keywords, and feeds back the search results to the user. For example, in a search engine for music scenes, the user can enter the song name, singer name or lyrics fragment, and the search engine feeds back the results according to the input content, including song links, singer information, full lyrics, etc.

[0004] However, by relying solely on keyword matching, the search engine will return all content containing the keyword, ignoring the user's intention, resulting in a large amount of irrelevant content in the search results and low search result accuracy. Summary of the invention

[0005] The embodiments of the present application provide a content search method, apparatus, device, medium and program product, which can improve the accuracy of search results. The technical solution is as follows:

[0006] In one aspect, a content search method is provided, the method comprising:

[0007] receiving a search request, wherein the search request includes a first search content, and the first search content is a search condition for data search;

[0008] Acquire associated search content corresponding to the first search content, wherein the associated search content is search content whose search trigger interval with the first search content is within a preset time range within a historical time period, and the first search content and the associated search content meet a preset association condition;

[0009] The first search content and the associated search content are searched to obtain search results.

[0010] In another aspect, a content search device is provided, the device comprising:

[0011] A receiving module, configured to receive a search request, wherein the search request includes a first search content, and the first search content is a search condition for data search;

[0012] an acquisition module, configured to acquire associated search content corresponding to the first search content, wherein the associated search content is search content whose search trigger interval with the first search content is within a preset time range within a historical time period, and the first search content and the associated search content meet a preset association condition;

[0013] The search module is used to search the first search content and the associated search content to obtain search results.

[0014] In an optional embodiment, the acquisition module further includes:

[0015] an acquisition unit, configured to acquire candidate associated content corresponding to the first search content, wherein the candidate associated content refers to at least one search content whose search triggering interval duration with the first search content is within the preset duration range within the historical time period;

[0016] A calculation unit, configured to calculate a search relevance between the candidate related content and the first search content;

[0017] The determining unit is configured to determine the target content as the associated search content if there is a target content in the candidate associated content and the search association degree corresponding to the target content meets the preset association condition.

[0018] In an optional embodiment, the acquisition unit is also used to acquire a search record corresponding to the first search content within the historical time period, the search record including a record generated when the first search content is received within the historical time period; and acquire the candidate associated content corresponding to the first search content from the search record.

[0019] In an optional embodiment, the search relevance includes a relevance search rate;

[0020] The calculation unit is further used to calculate the frequency of occurrence of the i-th search content in the search record for the i-th search content in the candidate associated content, and obtain an associated search rate corresponding to the i-th search content, where i is a positive integer.

[0021] In an optional embodiment, the search relevance includes content relevance;

[0022] The calculation unit is also used to calculate the i-th content correlation between the i-th search content and the first search content for the i-th search content in the candidate associated content, and the i-th content correlation is used to indicate the degree of correlation between the i-th search content and the first search content at the content topic and semantic levels.

[0023] In an optional embodiment, the calculation unit is further used to calculate, for the i-th search content in the candidate associated content, an i-th edit distance and an i-th set similarity between the i-th search content and the first search content, the i-th edit distance referring to the number of single-character editing operations required to convert the first search content into the i-th search content, and the i-th set similarity referring to the ratio of the intersection and the union between the first search content and the i-th search content; obtain the content topic of the i-th search content and the content topic of the first search content; calculate the i-th topic correlation between the i-th search content and the first search content, the i-th topic correlation being used to indicate whether the content topic of the i-th search content and the content topic of the first search content meet a preset correlation condition; obtain the i-th content correlation based on the i-th edit distance, the i-th set similarity and the i-th topic correlation.

[0024] In an optional embodiment, the determining unit is further configured to sort the candidate associated contents based on the search relevance, and determine the top j ranked contents among the candidate associated contents as the associated search contents, where j is a positive integer.

[0025] In an optional embodiment, the associated search content includes at least one sub-content;

[0026] The search module is further used to obtain the search correlation between the at least one sub-content and the first search content respectively, and the search correlation is used to describe the degree of correlation between the sub-content and the first search content; sort the at least one sub-content based on the search correlation to obtain a sorting result; search the first search content and the at least one sub-content to obtain a first search result corresponding to the first search content and a second search result corresponding to the at least one sub-content; and integrate the first search result and the second search result based on the sorting result to obtain the search result.

[0027] On the other hand, a computer device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement a content search method as described in any of the above-mentioned embodiments of the present application.

[0028] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement a content search method as described in any of the above-mentioned embodiments of the present application.

[0029] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes any of the content search methods described in the above embodiments.

[0030] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least:

[0031] Perform association analysis and historical tracing on the search content included in the search request to obtain the corresponding associated search content. The associated search content can reflect the user's satisfaction when receiving feedback on the same search content in the historical time period, and obtain the user's true intention when searching based on the first search content. On the basis of the original search content, continue to add the results obtained when searching for the associated search content, and feedback more accurate and comprehensive search results to the user with a full understanding of the user's intention, thereby improving the accuracy of the user's search based on the search content. In addition, to a certain extent, it reduces the frequency of search events in which users search again in a short period of time due to dissatisfaction with the search results, thereby improving the efficiency of user searches. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 is a schematic diagram of a content search system provided by an exemplary embodiment of the present application;

[0034] Figure 2 is a flow chart of a content search method provided by an exemplary embodiment of the present application;

[0035] Figure 3 is a flow chart of a method for simultaneously determining search relevance based on associated search rate and content relevance proposed by an exemplary embodiment of the present application;

[0036] Figure 4 is a flow chart of a method for obtaining search results provided by an exemplary embodiment of the present application;

[0037] Figure 5 is a structural block diagram of a content search device provided by an exemplary embodiment of the present application;

[0038] Figure 6 is a structural block diagram of a content search device provided by another exemplary embodiment of the present application;

[0039] Figure 7 It is a structural block diagram of a computer device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0040] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0041] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0042] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0043] It should be noted that the information and data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0044] It should be understood that, although the terms first, second, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first parameter may also be referred to as the second parameter, and similarly, the second parameter may also be referred to as the first parameter. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0045] First, a brief introduction to the nouns involved in the embodiments of the present application is given:

[0046] Associated search content: The associated search content corresponding to the first search content refers to the search content whose search trigger interval duration is within a preset duration range from the first search content within a historical time period.

[0047] Within a historical time period, a user sends a first search request containing the first search content to a device or application program capable of providing a data search function (such as a search engine or application program deployed on the server side or terminal), and the first search content included in the search request is used as a search condition to obtain search results. The user continues to send a second search request within a preset duration range after sending the first search request, and the search content included in the second search request is the associated search content.

[0048] That is to say, the associated search content corresponding to the first search content refers to the search condition input by the user when performing an associated search on the first search content.

[0049] The associated search content can reflect the user's satisfaction with the search results obtained from the first input search content and the user's true intention when performing data search based on the search content.

[0050] For example, the user inputs the text content A "The Wheat Waves Blowing in the Wind" as the first search content in a music scene search engine, and the content feedback by the search engine is as follows: Song list A with the song named "The Wheat Waves Blowing in the Wind", and song list A contains at least one song with the same name as "The Wheat Waves Blowing in the Wind".

[0051] The user continues to input the text content B "The Wheat Waves Blowing in the Wind - Lyrics" as the second search content within a preset duration range (for example, within 1 minute) after inputting the text content A, and the content feedback by the search engine is as follows: Song list B with the lyrics containing "The Wheat Waves Blowing in the Wind", and song list B contains at least one song, and the lyrics of the songs all contain "The Wheat Waves Blowing in the Wind".

[0052] At this time, the second search content is the associated search content of the first search content.

[0053] If the user continues to input the text content C "The Wheat Waves Blowing in the Wind - Lyrics - Female Singer" as the third search content after inputting the text content B, and the time duration between the moment of inputting the text content C and the moment of inputting the text content A still belongs to the preset duration range, the content feedback by the search engine is as follows: Song list C with the lyrics containing "The Wheat Waves Blowing in the Wind", and song list C contains at least one song sung by the same or different female singers, and the lyrics of the songs all contain "The Wheat Waves Blowing in the Wind".

[0054] At this time, the third search content is also the associated search content of the first search content.

[0055] Query: refers to the search content contained in the search request sent by the user, which is used to query and search for specific data or information. For example, the first search content can be called Query1, and the related search content can be called Query2.

[0056] Query2 is a query / search content relative to Query1. Query2 may refer to multiple search contents associated with Query1. For example, the second search content and the third search content mentioned above may both be referred to as Query2.

[0057] Recent N Secondary Search Rate / Associated Search Rate: The associated search rate between the associated search content and the first search content is used to indicate the frequency of users continuing to enter associated search content after entering the first search content in the historical time period, where the historical time period contains the records generated when the user searched for the first search content for the last N times. N is a positive integer and the value of N can be arbitrary.

[0058] It is worth noting that the "secondary search" mentioned in this application refers to the search event that continues within a preset time range for the search content input for the first time, and the number of search events can be arbitrary.

[0059] Usually, there are multiple users using the same search engine or the same search service, and the search contents included in the associated search events for the first search content performed by different users in the historical time period can all be used as the associated search contents of the first search content.

[0060] For example, the search records of the last 100 times for the first search content received by a search engine are obtained, and the search records include the search records generated when 100 users input the first search content a total of 100 times, among which 3 users continued to input the following content within a preset time range after inputting the first search content: the first user inputted search content 1 and search content 2, the second user inputted search content 1 and search content 3, and the third user inputted search content 2 and search content 4.

[0061] Then, the associated search contents include search content 1, search content 2, search content 3, and search content 4.

[0062] The process of calculating the correlation search rate between each associated search content and the first search content is as follows:

[0063] If search content 1 appears twice in the search record, the corresponding associated search rate of search content 1 is 2 / 100=2%; if search content 2 appears twice in the search record, the corresponding associated search rate of search content 2 is 2 / 100=2%; if search content 3 appears once in the search record, the corresponding associated search rate of search content 3 is 1 / 100=1%; if search content 4 appears once in the search record, the corresponding associated search rate of search content 4 is 1 / 100=1%.

[0064] As a key tool for Internet information retrieval, search engines play the role of a bridge connecting users with the ocean of network information. Search engines can index the massive multimedia content (including text, pictures, videos, audio and other forms of multimedia content) on the Internet or in databases based on search conditions, filter out the most relevant results and present the results to users.

[0065] The design of search engines aims to simplify the user's operation process. By providing a user-friendly input interface, users can easily enter query requests and efficiently retrieve and locate the information of their interest.

[0066] For example, using a search engine in a music scenario, a user can send a query instruction to the search engine by entering information such as the name of the song, the name of the singer, or part of the lyrics.

[0067] After receiving these keywords, the search engine will match them in its huge database and eventually return relevant search results including song links, detailed introduction of singers, complete lyrics, etc.

[0068] However, current search engine technologies often rely solely on keyword matching mechanisms when processing user queries.

[0069] Although this method can quickly return content containing specific keywords, due to the lack of in-depth understanding of the user's query intent, the search results are often filled with a large amount of information that is irrelevant to the user's actual needs. As a result, the accuracy of the search results is low, and the user may need to conduct a secondary search (that is, a related search) to obtain search results that better meet their expectations.

[0070] The present application provides a content search method, which can, after receiving a search request input by a user, perform association analysis and historical backtracking on a first search content contained in the search request to obtain associated search content corresponding to the first search content. The associated search content can reflect the probability of a secondary search event occurring when the user performs a data search on the first search content within a historical time period, and can reflect the user's actual search intention.

[0071] By rewriting (e.g., splicing, overlaying) the first search content originally contained in the search request through associated search content, and using the associated search content and the first search content together as search conditions to obtain corresponding search results and feed them back to the user, the efficiency of user search and the accuracy of search results can be improved.

[0072] The method provided in the present application can solve the problem in the related art that when the search results are fed back only by keyword matching, the search results may contain a large amount of irrelevant content.

[0073] Next, the content search system involved in the embodiment of the present application is described. For illustration, please refer to Figure 1 The system involves a first user terminal 110 and a server 120, and there is a communication connection between the first user terminal 110 and the server terminal 120.

[0074] The server 120 can provide a data search service. For example, a search application or a search engine is deployed in the server 120. The first user terminal 110 can provide a user with an interface entry for invoking the data search service.

[0075] In some embodiments, a search application or a search engine is deployed in the first user terminal 110, which can directly provide data search services to users.

[0076] The first user terminal 110 displays a search interface, which can receive the first search content input by the user. Exemplarily, the first search content may be in the form of text, audio, image, video, animation, or a file in a specific format.

[0077] After receiving the first search content, the first user terminal 110 generates a search request and sends the search request to the server 120 .

[0078] The search request includes a first search content, and the search request is used to request the server 120 to perform a data search based on the first search content, obtain search results from a database corresponding to the server 120, and feed back the search results to the user.

[0079] After receiving the first search content, the server 120 obtains the search records generated by all the user terminal groups 130 that have called the data search service to search for the first search content within the historical time period based on the first search content, and determines the associated search content corresponding to the first search content based on the associated search events in the search records.

[0080] The associated search content and the first search content meet a preset association condition.

[0081] Exemplarily, the search record includes records corresponding to the most recent 100 search events in which the user terminal group 130 invoked the data search service for the first search content, and the search record indicates that the user terminal group 130 includes search records of 90 user terminals in total.

[0082] That is, 90 users searched for the first search content 100 times in total.

[0083] Among them, there are 10 user terminals in the user terminal group 130 that performed related searches within a preset time range (for example, within 3 minutes) after searching for the first search content. All second search contents corresponding to the related searches and the number of occurrences are as follows: search content A appeared 10 times, search content B appeared 5 times, search content C appeared 8 times, and search content D appeared 5 times.

[0084] Exemplarily, the associated search content and the first search content satisfying a preset association condition means that the second search content ranked first n in search association degree with the first search content is determined as the associated search content, where n is a positive integer.

[0085] Among them, the search relevance includes the associated search rate and content relevance.

[0086] 1. Calculate the correlation search rate between each second search content and the first search content:

[0087] The associated search rate corresponding to search content A is 10 / 100=10%, the associated search rate corresponding to search content B is 5 / 100=5%, the associated search rate corresponding to search content C is 8 / 100=8%, and the associated search rate corresponding to search content D is 5 / 100=5%.

[0088] 2. Calculate the content relevance between each second search content and the first search content:

[0089] The subject content of each second search content and the subject content of the first search content are obtained.

[0090] The subject content is a pre-set tag content. The pre-set tag library is indexed according to the first search content and the second search content, so as to determine the subject content of the second search content and the subject content of the first search content.

[0091] 2.1 Determine the topic relevance value corresponding to each second search content based on the topic content.

[0092] For example, when the subject content of the second search content is the same as the subject content of the first search content, the subject relevance value is determined to be 1; when the subject content of the second search content is different from the subject content of the first search content, the subject relevance value is determined to be 0.

[0093] 2.2 Calculate the edit distance between each second search content and the first search content, where the edit distance refers to the number of single-character edit operations required to convert the first search content into the second search content.

[0094] For example, converting search content A (containing character 1, character 2, character 3 in sequence) into the first search content (containing character 2, character 1, character 3 in sequence) requires two operations (the operation of converting character 1 into character 2, and the operation of converting character 2 into character 1), and the edit distance is 2.

[0095] 2.3 Calculate the set similarity (Jaccard similarity) between each second search content and the first search content. The set similarity refers to the ratio of the intersection and union between the first search content and the second search content.

[0096] For example, the first search content contains 3 characters and the second search content contains 3 characters. The first search content is regarded as set A and the second search content is regarded as set B.

[0097] The intersection of set A and set B refers to a set of characters that appear in both the first search content and the second search content, and the union of set A and set B refers to all sets of characters that appear in the first search content and the second search content.

[0098] The intersection of set A and set B contains 2 characters, and the union of set A and set B contains 4 characters. The set similarity is: the ratio of the number of characters contained in the intersection to the number of characters contained in the union, 2 / 4=0.5.

[0099] For each second search content, the preset first weight, second weight and third weight are obtained, and the weighted operation values ​​between the topic relevance value, edit distance and set similarity are calculated to obtain the content relevance between each second search content and the first search content.

[0100] For example, taking search content A as an example, the topic relevance of search content A is 0, the edit distance is 2, the set similarity is 0.5, the preset first weight is 1, the second weight is 0.5 and the third weight is 0.5, then the content relevance between search content A and the first search content is 1*0+0.5*2+0.5*0.5=1.25.

[0101] For each second search content, the preset fourth weight and fifth weight are obtained, and the weighted operation value between the content relevance and the associated search rate is calculated based on the fourth weight and the fifth weight to obtain the search relevance between each second search content and the first search content.

[0102] For example, taking search content A as an example, the associated search rate corresponding to search content A is 10%, the content correlation between search content A and the first search content is 1.25, the preset fourth weight is 6, and the fifth weight is 4, then the search correlation between search content A and the first search content is 10%*6+1.25*4=5.6.

[0103] It is worth noting that the values ​​of the first weight, the second weight, the third weight, the fourth weight, and the fifth weight can be arbitrary.

[0104] Exemplarily, based on the above example content, the search relevances between search content A, search content B, search content C and search content D and the first search content are determined as follows: 5.6, 8.0, 9.1, 2.2, n is 1, and search content C with the highest search relevance is determined as the associated search content of the first search content.

[0105] After obtaining the associated search content corresponding to the first search content, a search is performed based on the first search content and the associated search content to obtain search results, wherein the search results include the results of searching the first search content and the results of searching the associated search content.

[0106] After obtaining the search results, the server 120 sends the search results to the first user terminal 110 , which presents them to the user.

[0107] The devices involved in the above-mentioned first user terminal 110 and user terminal group 130 can be various forms of terminal devices such as mobile phones, tablet computers, desktop computers, portable laptops, smart TVs, car terminals, smart home devices, etc., and the embodiments of the present application are not limited to this.

[0108] It is worth noting that the above-mentioned server 120 can 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, content delivery networks (CDN), as well as big data and artificial intelligence platforms.

[0109] Among them, cloud technology refers to a hosting technology that unifies hardware, software, network and other resources in a wide area network or local area network to achieve data computing, storage, processing and sharing. Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on the cloud computing business model. It can form a resource pool that can be used on demand and is flexible and convenient. Cloud computing technology will become an important support.

[0110] The backend services of technical network systems require a large amount of computing and storage resources, such as video websites, image websites and more portal websites. With the rapid development and application of the Internet industry, each item may have its own identification mark in the future, and all of them need to be transmitted to the backend system for logical processing. Data of different levels will be processed separately. All kinds of industry data require strong system backend support, which can only be achieved through cloud computing.

[0111] In some embodiments, the server 120 may also be implemented as a node in a blockchain system.

[0112] In combination with the above-mentioned noun introduction and application scenarios, the content search method provided by the present application is described. The method can be executed by a server or a terminal, or can be executed by a server and a terminal together. In the embodiment of the present application, the method is described by taking the execution of the server as an example. Figure 2 As shown, Figure 2 1 is a flow chart of a content search method provided by an exemplary embodiment of the present application. The method includes the following steps.

[0113] Step 210: Receive a search request.

[0114] An application program capable of providing data search service is run in the server and the terminal / a search engine is deployed thereon. A user inputs a first search content through an interface entrance of the data search service provided by the terminal to invoke the data search service.

[0115] After receiving the first search content, the terminal generates a search request based on the first search content. The search request includes the first search content. The first search content is a search condition for data search. The search condition is used for matching in a database corresponding to the data search service.

[0116] The expression form of the first search content includes, but is not limited to: text content, audio content, picture content, video content, animation content, file content in a specific format, etc.

[0117] The information contained in the first search content can be arbitrary. Data search services in different application fields correspond to different databases. The databases contain various types of content and can provide users with corresponding search results based on the first search content input by the user.

[0118] Optionally, the data search service can be applied to areas including but not limited to the following:

[0119] (1) Music: Data search services can be used to retrieve information such as songs, artists, and albums.

[0120] (2) Financial field: Data search services are used in financial data retrieval, market analysis, and investment decision-making.

[0121] (3) Academic field: Data search services are used in literature retrieval, academic research, and knowledge management.

[0122] (4) Medical field: Data search services are used in medical literature retrieval, disease diagnosis, and drug search.

[0123] (5) Manufacturing industry: Data search services are used in production process monitoring, quality control, and fault diagnosis.

[0124] (6) Media and entertainment: Data search services are used for content retrieval, copyright management, and user recommendations.

[0125] (7) Education and training: Data search services are used in learning resource retrieval, course management, and learning analysis.

[0126] For example, in a data search service in the field of music, according to the presentation form of the first search content, the first search content may include the following information:

[0127] 1. Text content: The first search content is the text content containing lyrics / song title / singer name / song theme / multimedia work (the multimedia work contains song content) name.

[0128] For example, the user inputs the text "name of singer A - song" and expects to search for music work information related to singer A.

[0129] 2. Audio content: The first search content is audio content containing songs / instrumental music / voice clips.

[0130] For example, a user uploads audio content obtained when recording a song "Sunny Day" and expects to search for music work information related to the song "Sunny Day".

[0131] 3. Picture content: The first search content is the picture content containing an album cover picture / a singer's photo / a promotional picture of a film or TV work.

[0132] For example, a user uploads the cover image of singer A's first album, hoping to search for music information related to the album.

[0133] 4. Video content: The first search content is video content containing a certain song MV (Music Video) clip / a certain film and television work clip / a certain music performance clip.

[0134] For example, a user uploads a video clip containing a recording of film and television work A, and expects to search for music work information related to the film and television work A.

[0135] 5. Animation content: The first search content is animation content containing animation clips or dynamic graphics related to a certain musical work.

[0136] For example, a user uploads an animation clip related to animation work A, and expects to search for music work information related to the animation work.

[0137] 6. File content in a specific format: The first search content is a file in a specific format, and the file content in this format can be parsed to obtain relevant information.

[0138] For example, a user uploads a music file in MP3 format (an audio encoding format) and expects to search for information related to the music file.

[0139] For example, in a data search service in the field of education and training, according to the presentation form of the first search content, the first search content may include the following information:

[0140] 1. Text content: The first search content is the text content containing subject name / textbook name / teacher name / university name.

[0141] For example, the user enters "tutoring institution A-high mathematics" and expects to search for high mathematics course content published by tutoring institution A.

[0142] 2. Audio content: The first search content is the audio content containing course recordings / listening test content.

[0143] For example, a user uploads audio content containing an English listening test segment and expects to search for complete listening audio content corresponding to the listening segment.

[0144] 3. Picture content: The first search content is the picture content containing a textbook cover / a teacher's photo / a university emblem / a logo of an educational institution.

[0145] For example, a user uploads a cover image of a university advanced mathematics textbook, hoping to search for course content corresponding to the advanced mathematics textbook.

[0146] 4. Video content: The first search content is video content containing clips of a teacher’s teaching / clip clips of an online course / clip clips of a school’s public lecture.

[0147] For example, a user uploads a video clip of an advanced mathematics online course and expects to search for online course content related to the video clip.

[0148] 5. Animation content: The first search content includes the animation content mentioned in the teaching courseware / teaching courses.

[0149] For example, a user uploads a playable animation content included in a teaching courseware of a high-level mathematics course, and hopes to search for content related to the teaching courseware.

[0150] 6. File content in a specific format: The first search content is a file in a specific format, and the file content in this format can be parsed to obtain relevant information.

[0151] For example, a user uploads a teaching courseware in the format of an e-book and expects to search for content related to the teaching courseware.

[0152] After generating the search request, the terminal sends the search request to the server, and the server performs a data search on the first search content included in the search request.

[0153] Step 220: Obtain the associated search content corresponding to the first search content.

[0154] The associated search content is the search content whose search triggering interval with the first search content is within a preset time range within a historical time period, and the first search content and the associated search content meet the preset association condition.

[0155] The historical time period refers to a time period with a preset time interval from the current moment. Methods for determining the preset time interval include but are not limited to the following.

[0156] (1) The preset time interval is a specified time interval, the moment when the current search request is received is determined as the current moment, and the time period from the current moment to the specified time interval is the historical time period. For example, the current moment is 12:00:00 on January 20, 2020, and the specified time interval is 240 hours, then the historical time period is the time period between 12:00:00 on January 10, 2020 and 12:00:00 on January 20, 2020.

[0157] (2) Determine the preset time interval based on the most recent N search records generated, obtain the N search records generated before the current time, and determine the historical time period based on the time corresponding to the earliest generated record in the search records, where N is a positive integer. For example, if N is 100, obtain the 100 search records generated before the current time, and determine the historical time period with the time corresponding to the earliest generated record among the 100 search records (12:00:00 on January 10, 2020) as the starting point and the current time (12:00:00 on January 20, 2020) as the end point.

[0158] (3) Determine a preset time interval based on the search records containing the first search content generated in the last M times, obtain multiple search records generated before the current moment, determine whether each search record contains the first search content, and obtain the candidate search records containing the first search content generated in the last M times. Determine the historical time period based on the time corresponding to the earliest generated record among the candidate search records, where M is a positive integer. For example, if M is 50, obtain 50 candidate search records containing the first search content generated before the current moment, and determine the historical time period with the time corresponding to the earliest generated record among the 50 candidate search records (12:00:00 on January 10, 2020) as the starting point and the current moment (12:00:00 on January 20, 2020) as the end point.

[0159] In the historical time period, the user inputs a first search content into the terminal, the terminal generates a first search request including the first search content and sends the first search request to the server, and the server obtains search results according to the first search content as a search condition.

[0160] The user continues to input the second search content within a preset time range after inputting the first search content, and the terminal generates a second search request containing the second search content and sends the second search request to the server. The second search content is the associated search content of the first search content.

[0161] That is, the associated search content corresponding to the first search content refers to the search condition input by the user when performing an associated search for the first search content.

[0162] The associated search content can reflect the user's satisfaction with the search results obtained from the search content entered for the first time, as well as the user's intention when searching for data based on the search content.

[0163] In order to enable users to obtain satisfactory search results when conducting their first search for the first search content, the server will conduct statistical analysis on the users' use of the data search function within a historical time period, and determine the related search content corresponding to the first search content, so that when users conduct data searches based on the first search content in the future, the server will immediately provide users with search results that meet their search intentions.

[0164] For example, there are usually multiple users using the same search engine or data search service. The server will collect statistics on the search events corresponding to all users who have searched for the first search content in the historical time period, and count whether the user continues to enter other search content within a preset time range after entering the first search content.

[0165] When there are other search contents, based on the frequency of occurrence of each search content in the other search contents and the relevance between each search content and the first search content in terms of content topic, associated search contents associated with the first search content are determined, and the user's search intent is fully understood based on the associated search contents and the first search content.

[0166] Optionally, the process of determining the associated search content includes the following steps:

[0167] Step S1: Acquire candidate related content corresponding to the first search content.

[0168] The candidate related content refers to at least one search content whose search triggering interval duration with the first search content is within a preset duration range within a historical time period.

[0169] A search record corresponding to the first search content within a historical time period is obtained, where the search record includes a record generated when the first search content is received within the historical time period.

[0170] Exemplarily, the historical time period refers to the time period corresponding to the most recent N times of receiving the search request containing the first search content, where N is a positive integer and N is a preset value.

[0171] Among them, each search record also includes the search content received within the preset time range after receiving the first search content sent by the user. That is, if the user continues to send other search content within the preset time range after sending the first search content, the search record also includes these search contents.

[0172] The candidate related content corresponding to the first search content is obtained from the search record, wherein the candidate related content may be from different users or from the same user.

[0173] Exemplarily, N is 100, and the historical time period is the time period corresponding to the last 100 search requests containing the first search content received: the time period between 00:00:00 on October 1, 2020 and 10:00:00 on October 3, 2020.

[0174] The 100 search requests containing the first search content come from 100 different users, and each user has searched for the first search content once in the historical time period.

[0175] The search records corresponding to the 100 search requests include 100 records, and each record includes, in addition to the first search content, the operation records and data generated within a preset time range after the user inputs the first search content.

[0176] The preset duration range is a duration range that starts from the moment when the first search content is input, and the duration range is m unit durations, where m is a positive integer.

[0177] After the user inputs the first search content through the interface provided by the terminal, when the terminal displays a prompt message indicating successful receipt, it means that the user has completed inputting the first search content.

[0178] For example, m is 3, the unit time is 1 minute, and the preset time range refers to within 3 minutes after the user inputs the first search content.

[0179] Among the 100 search records, there are 5 records indicating that the user continues to input candidate related content within a preset time range after inputting the first search content.

[0180] The triggering conditions for the "continue input" operation mentioned here include but are not limited to:

[0181] (1) Enter candidate related content in the input box provided in the terminal interface.

[0182] For example, after the user enters the first search content in the input box and clicks the send control, the first search content is sent to the server. Then, the user enters candidate related content in the input box within a preset time period. When there is candidate related content in the input box, it is considered that the user continues to "continue entering" within the preset time period after entering the first search content.

[0183] (2) Input candidate related content in the input box provided in the terminal interface, and send the candidate related content to the server.

[0184] For example, after the user enters the first search content in the input box and clicks the send control, the first search content is sent to the server. Then, within a preset time period, the user enters candidate related content in the input box and clicks the send control to send the candidate related content to the server. When the candidate related content exists in the input box and is successfully sent to the server, it is considered that the user continues to "continue entering" within the preset time period after entering the first search content.

[0185] For example, within 3 minutes after user 1 inputs the first search content, he continues to input search content A and search content B in the input order.

[0186] Within 3 minutes after user 2 inputs the first search content, he continues to input search content A and search content C in the order in which he inputs.

[0187] Within 3 minutes after user 3 inputs the first search content, he continues to input search content D, search content A, and search content B in the order of input.

[0188] Within 3 minutes after user 4 inputs the first search content, he continues to input search content C and search content A in the order in which he inputs.

[0189] Within 3 minutes after user 5 inputs the first search content, he continues to input search content B, search content C, and search content A in the order in which they were input.

[0190] Then, the candidate related contents include: search content A, search content B, search content C, and search content D.

[0191] Step S2: Calculate the search relevance between the candidate related content and the first search content.

[0192] The candidate associated content includes at least one search content, and the search relevance between each search content and the first search content is calculated respectively. The search relevance can characterize the degree of association between the candidate associated content and the first search content, and reflect the user's intention and search tendency when searching for the first search content.

[0193] Optionally, the search relevance includes a relevance search rate.

[0194] For the i-th search content in the candidate related content, the frequency of the i-th search content appearing in the search record is calculated to obtain the related search rate corresponding to the i-th search content, where i is a positive integer.

[0195] For example, the associated search rate corresponding to search content A is 5 / 100=5%, the associated search rate corresponding to search content B is 3 / 100=3%, the associated search rate corresponding to search content C is 3 / 100=3%, and the associated search rate corresponding to search content D is 1 / 100=1%.

[0196] Optionally, the search relevance includes content relevance.

[0197] For the ith search content in the candidate associated content, the ith content relevance between the ith search content and the first search content is calculated. The ith content relevance is used to indicate the degree of relevance between the ith search content and the first search content at the content theme and semantic level.

[0198] Exemplarily, a subject tag file is obtained, wherein the subject tag file contains a mapping relationship between a subject tag pre-annotated by the server for each search content and the search content, and the subject tag is used to indicate a content theme corresponding to each search content.

[0199] Obtaining the topic tag of the i-th search content and the topic tag of the first search content based on the topic tag file, calculating the similarity between the topic tags, and determining the content relevance between the i-th search content and the first search content based on the similarity, including the following two cases:

[0200] (1) Determine the value of content relevance based on whether the similarity between topic tags reaches a preset similarity threshold.

[0201] When the similarity reaches a preset similarity threshold, the content relevance value is determined to be a first value, and when the similarity does not reach the preset similarity threshold, the content relevance value is determined to be a second value, and the first value and the second value are different.

[0202] For example, the i-th search content is search content A.

[0203] The first search content is "Blue and White Porcelain - Live Video", and the theme tag of the first search content obtained from the theme tag file is: music scene.

[0204] The search content A is "cover-blue and white porcelain-live", and the theme tag of the search content A obtained from the theme tag file is: music live.

[0205] The preset similarity threshold is 0.7, the first value is 1, and the second value is 0.

[0206] Among them, since the corresponding topic tags of the two are the same, the similarity between the topic tags is determined to be 1, reaching the preset similarity threshold of 0.7, and the content relevance value corresponding to the search content A is determined to be the first value 1.

[0207] For example, the i-th search content is search content B.

[0208] The first search content is "Blue and White Porcelain - Live Video", and the theme tag of the first search content obtained from the theme tag file is: music scene.

[0209] The search content B is "name of singer A - blue and white porcelain - cover", and the theme tag of the search content B obtained from the theme tag file is: music cover.

[0210] The preset similarity threshold is 0.7, the first value is 1, and the second value is 0.

[0211] Among them, since the corresponding topic labels of the two are not exactly the same, the two topic labels are input into a pre-trained language model, which can convert the input content into feature vectors and calculate the cosine similarity between the feature vectors. The cosine similarity is used as the similarity between the topic labels.

[0212] Exemplarily, the cosine similarity between the model output topic tags is 0.6, which does not reach the preset similarity threshold of 0.7, and it is determined that the content relevance value corresponding to the search content B is the second value 0.

[0213] (2) The similarity value between the topic tag of the i-th search content and the topic tag of the first search content is determined as the value of the content relevance.

[0214] For example, if the similarity between the topic tag of search content A and the topic tag of the first search content is 1, then the content relevance is 1.

[0215] For example, if the similarity between the topic tag of search content B and the topic tag of the first search content is 0.6, then the content relevance is 0.6.

[0216] Optionally, the search relevance between the candidate related content and the first search content is determined based on at least one of the related search rate and the content relevance, including the following situations:

[0217] (1) Determine search relevance based only on the relevance search rate.

[0218] For example, the value of the correlation search rate between the ith search content and the first search content is directly used as the value of the search correlation degree.

[0219] (2) Determine search relevance based only on content relevance.

[0220] For example, the value of the content relevance between the ith search content and the first search content is directly used as the value of the search relevance.

[0221] (3) Determine the search relevance based on the associated search rate and content relevance at the same time.

[0222] For example, a preset weight value is obtained, and a weighted operation is performed on the associated search rate and the content relevance to obtain the search relevance.

[0223] Exemplarily, a first preset weight and a second preset weight are obtained, a first product is obtained based on the product between the first preset weight and the associated search rate, a second product is obtained based on the product between the second preset weight and the content relevance, and a search relevance is obtained based on the sum of the first product and the second product.

[0224] For example, the first preset weight value is 0.4, the second preset weight value is 0.6, the associated search rate is 5%, and the content relevance value is 1, then the search relevance is 0.4*5%+0.6*1=0.62.

[0225] Step S3: If there is a target content in the candidate related content whose search relevance meets the preset relevance condition, the target content is determined as the related search content.

[0226] Optionally, the candidate associated contents are ranked based on the search relevance, and the top j ranked contents among the candidate associated contents are determined as associated contents, where j is a positive integer.

[0227] The ranking results are ranked from high to low according to the search relevance of the candidate related contents.

[0228] For example, the search relevances of search contents A, B, C, and D are 0.62, 0.55, 0.8, and 0.2, respectively.

[0229] The candidate related contents are sorted based on the search relevance, and the sorting results are as follows: the search content C is numbered 1, the search content A is numbered 2, the search content B is numbered 3, and the search content D is numbered 4.

[0230] When j is 2, the first two contents in the ranking are determined as related search contents, that is, search content C and search content A are related search contents of the first search content.

[0231] Step 230: Search the first search content and the associated search content to obtain search results.

[0232] The search results include a first result of searching the first search content and a second result of searching the associated search content.

[0233] When the server feeds back search results to the terminal, the search results are displayed in a list format. The search results include a first result and a second result arranged in order. The first result includes at least one multimedia content, and the second result also includes at least one multimedia content.

[0234] The types of multimedia content include but are not limited to text content, audio content, picture content, video content, animation content, file content in a specific format, etc. When there are multiple multimedia contents in the search results, the arrangement order is determined based on the search conditions.

[0235] Optionally, browsing data corresponding to each multimedia content in the first result and the second result is obtained, and the browsing data includes browsing frequency, browsing times and complete browsing rate of the multimedia content.

[0236] The browsing frequency refers to the proportion of the browsing time of the multimedia content within a unit time range.

[0237] The number of views refers to the total number of views of the multimedia content by all users who call the data search service.

[0238] The complete browsing rate refers to the ratio between the number of times the multimedia content is completely browsed and the number of times the multimedia content is browsed.

[0239] An arrangement order of each multimedia content is determined based on the browsing data.

[0240] Exemplarily, all multimedia contents in the first result and the second result are regarded as a content set, and for the i-th multimedia content in the content set, the browsing frequency, number of views and complete browsing rate in the browsing data of the i-th multimedia content are weightedly calculated to obtain the ranking value of the i-th multimedia content.

[0241] All multimedia contents are sorted based on the arrangement values ​​to obtain a sorting result, and the multimedia contents are displayed in the search results (first result and second result) based on the sorting result.

[0242] For example, for the i-th multimedia content, the browsing duration within the unit duration range (5 minutes) is the first duration (1 minute), and the browsing frequency is 1 / 5=20%.

[0243] The number of views is 500, the number of complete views is 250, and the complete view rate is 250 / 500=50%.

[0244] The weights assigned to browsing frequency, browsing times and complete browsing rate are 1, 0.01 and 1 respectively, and the ranking value obtained by weighted calculation is 1*20%+0.01*500+1*50%=5.7.

[0245] The first result and the second result contain 5 multimedia contents in total, namely multimedia content 1, multimedia content 2, multimedia content 3, multimedia content 4, and multimedia content 5.

[0246] Based on the above example, the arrangement values ​​corresponding to each multimedia content are calculated to be 6.7, 5.7, 9, 7.2, and 3.4, and the ranking results are: multimedia content 3, multimedia content 4, multimedia content 1, multimedia content 2, and multimedia content 5.

[0247] In the search results, the five multimedia contents are displayed in sequence according to the sorting results and the specified direction, and the specified direction can be arbitrary.

[0248] The specified direction is the browsing direction when the user browses the multimedia content in the search results through the terminal. The specified direction is the direction within the plane where the terminal screen is located.

[0249] For example, specify the direction as from left to right / from top to bottom / from top left to bottom right, etc.

[0250] In summary, the content search method provided by the present application performs association analysis and historical tracing on the search content contained in the search request to obtain corresponding associated search content. The associated search content can reflect the user's satisfaction when receiving feedback on the same search content in a historical time period, and obtain the user's true intention when searching based on the first search content. On the basis of the original search content, the results obtained when searching for the associated search content are continued to be added, and more accurate and comprehensive search results can be fed back to the user with a full understanding of the user's intention, thereby improving the accuracy of the user's search based on the search content. In addition, the frequency of search events in which users search again in a short period of time due to dissatisfaction with the search results is reduced to a certain extent, thereby improving the efficiency of user searches.

[0251] In some embodiments, the first search content and the second search content are not related in terms of semantics or content subject, but are more likely to appear in the same context or be mentioned at the same time in a short period of time. This situation is accidental and will affect the server's judgment on whether there is a correlation between the search contents, resulting in inaccurate search results.

[0252] For example, there is a first character in a certain film or television work, and when the first character appears, there is background music accompanying him, and the background music is a song sung by a first singer. After the film or television work is broadcast, the name of the first character and the name of the background music appear at the same time at a high frequency (reaching a preset threshold).

[0253] After the user inputs the first search content "name of the first character + appearance music" using the search engine, he continues to input the second search content "name of the first singer + song title".

[0254] After the broadcast of the film or TV work, the frequency of the simultaneous appearance of the first character's name and the background music's song title decreased (did not reach the preset threshold), indicating that in the future, when the user enters the first character's name, the probability of seeing the background music's song title in the search results is low.

[0255] In order to improve the accuracy of search results and avoid accidental problems that occur after the user enters the first search content, so that the associated search content remains relevant to the first search content in terms of content subject and semantic level, step 220 can also be implemented as the following steps 310 to 340.

[0256] like Figure 3 As shown, Figure 3 It is a flow chart of a method for simultaneously determining search relevance based on associated search rate and content relevance, proposed by an exemplary embodiment of the present application.

[0257] Step 310: For the i-th search content in the candidate related content, the frequency of the i-th search content appearing in the search record is calculated to obtain the related search rate corresponding to the i-th search content.

[0258] Wherein, i is a positive integer.

[0259] For example, the search record includes the records corresponding to the most recent 100 search events in which users called the data search service for the first search content, and the search record indicates that there are 90 users who searched for the first search content in the historical time period, that is, 90 users searched for the first search content 100 times.

[0260] Among them, there are 10 users who performed related searches within a preset time range (for example, within 3 minutes) after searching the first search content, and the candidate related contents corresponding to the related searches include search content A, search content B, search content C, and search content D.

[0261] Among them, the number of times the candidate related content appears is as follows: search content A appears 10 times, search content B appears 5 times, search content C appears 8 times, and search content D appears 5 times.

[0262] The associated search rate corresponding to search content A is 10 / 100=10%, the associated search rate corresponding to search content B is 5 / 100=5%, the associated search rate corresponding to search content C is 8 / 100=8%, and the associated search rate corresponding to search content D is 5 / 100=5%.

[0263] Step 320: For the ith search content in the candidate associated content, calculate the ith content correlation between the ith search content and the first search content.

[0264] The i-th content relevance is used to indicate the relevance between the i-th search content and the first search content at the content topic and semantic levels.

[0265] 1. For the i-th search content in the candidate related content, calculate the i-th edit distance and the i-th set similarity between the i-th search content and the first search content.

[0266] Among them, the i-th edit distance refers to the number of single-character editing operations required to convert the first search content into the i-th search content, and the i-th set similarity refers to the ratio of the intersection and union between the first search content and the i-th search content.

[0267] Exemplarily, the first search content contains a plurality of first characters arranged in order, and the i-th search content contains a plurality of second characters arranged in order. For the m-th first character in the first search content, the number of operations required to convert it into the m-th second character in the i-th search content is counted, and the total number of operations required to convert all the characters in the first search content into the corresponding characters in the i-th search content in order is determined as the i-th edit distance.

[0268] Among them, when converting the first search content into the i-th search content, there are multiple conversion methods, and the types of editing operations included in each conversion method include but are not limited to:

[0269] (1) Conversion operation: converting one character into another different character;

[0270] (2) Insertion operation: inserting a character at any position in the search content;

[0271] (3) Deletion operation: deleting a character at any position in the search content, etc.

[0272] The one with the least number of operations in each conversion method is used as the target conversion method, and the edit distance corresponding to this target conversion method is the i-th edit distance.

[0273] For example, the first search content is "Unforgettable Tonight", which contains 4 first characters, and the i-th search content is "Farewell to Tonight", which contains 4 second characters.

[0274] When converting the first search content into the i-th search content, 2 operations are required: the operation of converting the character "难 (nán)" into the character "告 (gào)", and the operation of converting the character "忘 (wàng)" into the character "别 (bié)", so the edit distance is 2.

[0275] For example, the first search content is "Unforgettable Tonight", which contains 4 first characters, and the i-th search content is "Like Tonight Yesterday", which contains 6 second characters.

[0276] When converting the first search content into the i-th search content, 4 operations are required: the operation of converting the character "难 (nán)" into the character "如 (rú)", the operation of converting the character "忘 (wàng)" into the character "同 (tóng)", the operation of inserting the character "昨 (zuó)", and the operation of inserting the character "日 (rì)", so the edit distance is 4.

[0277] Exemplarily, the first search content contains 3 characters, and the i-th search content contains 3 characters. The first search content is regarded as the first set, and the i-th search content is regarded as the second set.

[0278] The intersection of the first set and the second set refers to the set of characters that appear in both the first search content and the i-th search content. The union of the first set and the second set refers to the set of all characters that appear in the first search content and the i-th search content.

[0279] The set similarity is: the ratio between the number of characters contained in the intersection and the number of characters contained in the union.

[0280] For example, the first search content is "Unforgettable Tonight", which contains 4 characters, and the i-th search content is "Farewell Tonight", which contains 4 characters.

[0281] The intersection of the first set and the second set contains 2 characters: "今" and "宵", and the union of the first set and the second set contains 6 characters: "难", "忘", "今", "宵", "告", "别". Then, the set similarity is: 2 / 6 = 33.3%.

[0282] II. Obtain the content theme of the i-th search content and the content theme of the first search content.

[0283] Among them, the theme content is the pre-set tag content. By indexing the pre-set tag library according to the first search content and the i-th search content, the theme content of the i-th search content and the theme content of the first search content can be determined.

[0284] Calculate the i-th theme relevance between the i-th search content and the first search content. The i-th theme relevance is used to indicate whether the content theme of the i-th search content and the content theme of the first search content meet the pre-set relevance conditions.

[0285] Exemplarily, when the theme content of the second search content is the same as the theme content of the first search content, it is determined that the theme relevance value is the first theme value; when the theme content of the second search content is different from the theme content of the first search content, it is determined that the theme relevance value is the second theme value.

[0286] The first theme value and the second theme value are pre-set different values.

[0287] For example, the first theme value is 1 and the second theme value is 0. Taking the i-th search content as search content A as an example for illustration.

[0288] The first search content is "Blue and White Porcelain - Live Video", and the theme content of the first search content obtained from the tag library is: Music Live.

[0289] The i-th search content is "cover-blue and white porcelain-live", and the subject content of the search content A obtained from the tag library is: music live.

[0290] Among them, since the corresponding subject contents of the two are the same, the subject correlation value is determined to be 1 for the first subject.

[0291] For example, the i-th search content is search content B.

[0292] The first search content is "Blue and White Porcelain - Live Video", and the subject content of the first search content obtained from the tag library is: Music Live.

[0293] The i-th search content is "name of singer A - blue and white porcelain - cover", and the subject content of search content B obtained from the tag library is: music cover.

[0294] Among them, since the corresponding subject contents of the two are different, the subject correlation value is determined to be 0 for the second subject.

[0295] In some embodiments, the value of the topic relevance may also be determined by calculating the similarity between the topic contents and based on whether the similarity between the topic contents reaches a preset similarity threshold.

[0296] When the similarity reaches a preset similarity threshold, the topic relevance value is determined to be a first topic value. When the similarity does not reach the preset similarity threshold, the topic relevance value is determined to be a second topic value. The first topic value and the second topic value are preset different values.

[0297] For example, the i-th search content is search content B.

[0298] The first search content is "Blue and White Porcelain - Live Video", and the subject content of the first search content obtained from the tag library is: Music Live.

[0299] The i-th search content is "name of singer A - blue and white porcelain - cover", and the subject content of search content B obtained from the tag library is: music cover.

[0300] Among them, since the corresponding subject contents of the two are not exactly the same, the two subject contents are input into a pre-trained language model. The pre-trained language model can convert the input content into a feature vector and calculate the cosine similarity between the two feature vectors, and use the cosine similarity as the similarity between the subject contents.

[0301] Exemplarily, the preset similarity threshold is 0.7, the first topic value is 1, and the second topic value is 0.

[0302] The cosine similarity between the model output topic contents is 0.6, which does not reach the preset similarity threshold of 0.7. It is determined that the topic relevance value corresponding to the i-th search content is 0, the second topic value.

[0303] Optionally, after obtaining the i-th edit distance, i-th set similarity, and i-th topic relevance corresponding to the i-th search content, the i-th content relevance is obtained based on the i-th edit distance, the i-th set similarity, and the i-th topic relevance.

[0304] Exemplarily, for the i-th search content, the preset first weight, second weight and third weight are obtained, and the weighted operation value between the topic relevance value, edit distance and set similarity is calculated to obtain the content relevance between the i-th search content and the first search content.

[0305] For example, taking the i-th search content as search content A, the topic relevance of search content A is 0, the edit distance is 2, the set similarity is 0.5, the preset first weight is 1, the second weight is 0.5 and the third weight is 0.5, then the content relevance between search content A and the first search content is 1*0+0.5*2+0.5*0.5=1.25.

[0306] It is worth noting that the above-mentioned step 310 and step 320 are two parallel steps, that is, step 310 can be executed first, and then step 320; or, step 320 can be executed first, and then step 310; or, step 310 and step 320 can be executed at the same time, which is not limited in this embodiment.

[0307] Step 330 : Perform a weighted operation on the associated search rate corresponding to the ith search content and the ith content relevance to obtain the search relevance between the ith search content and the first search content in the candidate associated content.

[0308] Optionally, for the i-th search content, the preset fourth weight and fifth weight are obtained, and the weighted operation value between the content relevance and the associated search rate is calculated based on the fourth weight and the fifth weight to obtain the search relevance between the i-th search content and the first search content.

[0309] For example, taking the i-th search content as search content A, the associated search rate corresponding to search content A is 10%, the content correlation between search content A and the first search content is 1.25, the preset fourth weight is 6, and the fifth weight is 4, then the search correlation between search content A and the first search content is 10%*6+1.25*4=5.6.

[0310] Exemplarily, based on the above example content, the search relevances between search content A, search content B, search content C, and search content D and the first search content are determined as follows: 5.6, 8.0, 9.1, and 2.2, respectively.

[0311] Step 340: after obtaining the search relevances corresponding to all search contents in the candidate related contents, the candidate related contents are sorted based on the search relevances, and the top j contents in the candidate related contents are determined as the related search contents.

[0312] Wherein, j is a positive integer.

[0313] Exemplarily, if j is 2, the candidate related contents are sorted from high to low based on the size of the search relevance, and the sorting results are: search content C (9.1), search content B (8.0), search content A (5.6), and search content D (2.2). The first two contents in the sorting results (i.e., search content C and search content B) are determined as related search contents.

[0314] To sum up, the method provided by the present application can ensure that the associated search content remains relevant to the first search content in terms of both content subject and semantic levels, thereby improving the accuracy of search results, avoiding the problem of low search result accuracy caused by users accidentally entering irrelevant content after entering the first search content, and improving users' search efficiency.

[0315] When the first search content is expanded and corresponding search results are obtained based on the associated search content obtained by the expansion and the first search content, if there are multiple results corresponding to the associated search contents in the search results, it is necessary to sort the search results corresponding to the multiple associated search contents, so that when the terminal feeds back the received search results to the user, the search results that best match the user's intentions can be displayed to the user first, thereby improving the search result feedback effect.

[0316] Step 230 can also be implemented as the following steps 410 to 440, such as Figure 4 As shown, Figure 4 It is a flow chart of a method for obtaining search results provided by an exemplary embodiment of the present application. The method includes the following steps, wherein the associated search content includes at least one sub-content.

[0317] Step 410: Obtain the search relevance between at least one sub-content and the first search content.

[0318] The search relevance is used to describe the relevance between the sub-content and the first search content.

[0319] Optionally, the search relevance includes associated search rate and content relevance.

[0320] The associated search rate corresponding to the i-th sub-content is used to indicate the frequency of a user continuing to input the i-th sub-content after inputting the first search content in a historical time period.

[0321] The content relevance corresponding to the i-th sub-content is used to indicate the relevance of the i-th sub-content to the first search content at the content theme and semantic level, where i is a positive integer.

[0322] Exemplarily, the associated search content includes four sub-contents, namely: search content A, search content B, search content C, and search content D.

[0323] The associated search rate corresponding to search content A is 10%, the associated search rate corresponding to search content B is 5%, the associated search rate corresponding to search content C is 8%, and the associated search rate corresponding to search content D is 5%.

[0324] The content relevance corresponding to search content A is 1.25, the content relevance corresponding to search content B is 1.2, the content relevance corresponding to search content C is 2.5, and the content relevance corresponding to search content D is 0.5.

[0325] The preset weights are obtained to perform weighted calculation on the associated search rate and the content relevance to obtain the search relevance, wherein the preset weights include a fourth weight and a fifth weight.

[0326] For the i-th sub-content, a first product is obtained based on the product of the fourth weight and the associated search rate corresponding to the i-th sub-content, a second product is obtained based on the product of the fifth weight and the content relevance corresponding to the i-th sub-content, and the search relevance corresponding to the i-th sub-content is obtained based on the sum of the first product and the second product.

[0327] For example, the fourth weight is 8 and the fifth weight is 2. The search relevance corresponding to search content A is 10%*8+1.25*2=3.3, the search relevance corresponding to search content B is 5%*8+1.2*2=2.8, the search relevance corresponding to search content C is 8%*8+2.5*2=5.64, and the search relevance corresponding to search content D is 5%*8+0.5*2=1.4.

[0328] Step 420: sort at least one sub-content based on the search relevance to obtain a sorting result.

[0329] At least one sub-content is sorted from high to low based on the search relevance, and the greater the search relevance, the higher the sequence number.

[0330] For example, in at least one sub-content, the search relevance corresponding to search content A is 3.3, the search relevance corresponding to search content B is 2.8, the search relevance corresponding to search content C is 5.64, and the search relevance corresponding to search content D is 1.4.

[0331] The corresponding sorting results are: (1) search content C; (2) search content A; (3) search content B; (4) search content D.

[0332] Step 430: Search the first search content and the at least one sub-content to obtain a first search result corresponding to the first search content and a second search result corresponding to the at least one sub-content.

[0333] Exemplarily, the search field corresponding to the data search service is the music field.

[0334] The first search content is "Sunny Day", and the first search result includes 4 multimedia contents arranged in a preset order: the song "Sunny Day" sung by the first singer, the song of the same name "Sunny Day" sung by the second singer, the first song whose lyrics include "Sunny Day", and the second song whose lyrics include "Sunny Day".

[0335] In at least one subcontent:

[0336] Search content C is "first singer + sunny day", and the second search result corresponding to search content C includes 2 multimedia contents arranged in order: the song "Sunny Day" sung by the first singer, and the third song sung by the first singer whose lyrics include "Sunny Day".

[0337] Search content A is "Sunny Day + Concert", and the second search result corresponding to search content A includes 1 multimedia content: the song "Sunny Day" sung live by the first singer at the concert.

[0338] Search content B is "first singer", and the second search result corresponding to search content B includes 2 multimedia contents arranged in order: the third song sung by the first singer whose lyrics include "Sunny Day", and the song "Sunny Day" sung by the first singer.

[0339] The search content D is "Sunny Day + Cover", and the second search result corresponding to the search content D includes 1 multimedia content: multimedia content of a third singer's cover of the first singer's song "Sunny Day".

[0340] Step 440: Integrate the first search result and the second search result based on the ranking result to obtain a search result.

[0341] Optionally, multiple multimedia contents in the first search result and the second search result are integrated based on a preset integration rule to obtain a search result.

[0342] The preset integration rules include but are not limited to the following:

[0343] (1) A sorting result of sorting at least one sub-content based on the search relevance is obtained by sequentially concatenating the first search result and the second search result to obtain a search result.

[0344] (2) Sorting the plurality of multimedia contents in the second search content based on the occurrence frequency and search relevance of each multimedia content in the second search result to obtain a second content sorting result, and integrating the second content sorting result and the first search result to obtain a search result.

[0345] Exemplarily, for integration rule (1), 4 multimedia contents in the first search result, 2 multimedia contents in the second search result corresponding to search content C, 1 multimedia content in the second search result corresponding to search content A, 2 multimedia contents in the second search result corresponding to search content B, and 1 multimedia content in the second search result corresponding to search content D are displayed in sequence according to the sorting results.

[0346] Since some multimedia contents overlap in the first search result and the second search result, the multimedia contents of the overlapping part are displayed only once in the search result based on the position where the multimedia contents first appear.

[0347] For example, for integration rule (1), ignoring overlaps, the search results display the following in order:

[0348] [1] The song "Sunny Day" sung by the first singer, [2] The song "Sunny Day" of the same name sung by the second singer, [3] The first song whose lyrics contain "Sunny Day", [4] The second song whose lyrics contain "Sunny Day", [5] The song "Sunny Day" sung by the first singer, [6] The third song whose lyrics contain "Sunny Day" sung by the first singer, [7] The song "Sunny Day" sung live by the first singer at a concert, [8] The third song whose lyrics contain "Sunny Day" sung by the first singer, [9] The song "Sunny Day" sung by the first singer,

[10] Multimedia content in which the third singer sings a cover of the first singer's song "Sunny Day".

[0349] Among them, if the multimedia contents with serial numbers [1], [5], and [9] overlap (that is, they are the same multimedia contents), then only the multimedia contents with serial number [1] will be displayed in the search results, and the multimedia contents with serial numbers [5] and [9] will not be displayed.

[0350] For example, for integration rule (1), considering the overlap, the search results display the following contents in order:

[0351] [1] The song "Sunny Day" sung by the first singer, [2] The song "Sunny Day" of the same name sung by the second singer, [3] The first song whose lyrics contain "Sunny Day", [4] The second song whose lyrics contain "Sunny Day", [6] The third song whose lyrics contain "Sunny Day" sung by the first singer, [7] The song "Sunny Day" sung live by the first singer at a concert, [8] The third song whose lyrics contain "Sunny Day" sung by the first singer,

[10] Multimedia content in which the third singer sings a cover of the first singer's song "Sunny Day".

[0352] Exemplarily, for integration rule (2), the occurrence frequency of each multimedia content in the second search content is determined.

[0353] Among them, the multimedia contents with serial numbers [5] and [9] overlap, so the multimedia contents with serial numbers [5] and [9] can be regarded as the same multimedia content, and their occurrence frequency is 2 / 6=33.3%.

[0354] The multimedia contents with serial numbers [6], [7], [8], and

[10] appear only once, and their frequency of appearance is 1 / 6=16.6%.

[0355] A weighted operation is performed based on the search relevance between the sub-content corresponding to each multimedia content in the second search result and the first search content, and the appearance frequency of each multimedia content in the second search result to obtain a content sequence value.

[0356] The plurality of multimedia contents in the second search result are sorted based on the content sequence number values ​​to obtain a content sorting result.

[0357] Optionally, for the i-th multimedia content in the second search result, the occurrence frequency of the i-th multimedia content is used as a weight, and the product obtained by multiplying it by the corresponding search relevance is used as the content sequence value.

[0358] For example, the content sequence number of the multimedia content with the sequence number [5] is: 33.3%*5.64=1.88;

[0359] The content sequence number of the multimedia content with sequence number [6] is: 16.6%*5.64=0.94;

[0360] The content sequence number of the multimedia content with sequence number [7] is: 16.6%*3.3=0.55;

[0361] The content sequence number of the multimedia content with sequence number [8] is: 16.6%*2.8=0.46;

[0362] The content sequence number of the multimedia content with sequence number [9] is: 33.3%*2.8=0.93;

[0363] The multimedia content with serial number

[10] has a content serial number value of: 16.6%*1.4=0.23.

[0364] The multimedia contents in the second search result are sorted based on the content sequence number values, and the content sorting results are as follows: multimedia contents with sequence numbers [5], [6], [9], [7], [8], and

[10] .

[0365] The first search result and the second search result are cross-integrated to obtain the following search results: multimedia content with serial numbers [5], [1], [6], [2], [9], [3], [7], [4], [8], and

[10] .

[0366] The cross integration refers to interlacing and arranging the multimedia contents in the first search result and the second search result, so that any two adjacent multimedia contents in the first 2n multimedia contents in the search result come from different search results.

[0367] The first search result includes n1 multimedia contents, and the second search result includes n2 multimedia contents, where n refers to a smaller value between n1 and n2, and n, n1, and n2 are all positive integers.

[0368] In some embodiments, if the multimedia contents numbered [1], [5], and [9] overlap, the multimedia contents are displayed only when they appear for the first time, and the search results are as follows: multimedia contents numbered [5], [6], [2], [3], [7], [4], [8], and

[10] .

[0370] In summary, the method provided by the present application can sort the search results obtained based on the first search content and the related search content respectively, and display the search results based on the sorting results. When the terminal feeds back the received search results to the user, the search results that best match the user's intention can be displayed to the user first, thereby improving the search result feedback effect.

[0371] Figure 5 is a structural block diagram of a content search device provided by an exemplary embodiment of the present application, such as Figure 5 As shown, the device includes the following parts.

[0372] A receiving module 510 is configured to receive a search request, wherein the search request includes a first search content, and the first search content is a search condition for data search;

[0373] The acquisition module 520 is used to acquire the associated search content corresponding to the first search content, wherein the associated search content is the search content whose search trigger interval with the first search content is within a preset time range within a historical time period, and the first search content and the associated search content meet the preset association condition;

[0374] The search module 530 is configured to search the first search content and the associated search content to obtain search results.

[0375] In an optional embodiment, if Figure 6 As shown, the acquisition module 520 further includes:

[0376] An acquisition unit 521 is configured to acquire candidate associated content corresponding to the first search content, wherein the candidate associated content refers to at least one search content whose search triggering interval duration with the first search content is within the preset duration range within the historical time period;

[0377] A calculation unit 522, configured to calculate a search relevance between the candidate related content and the first search content;

[0378] The determining unit 523 is configured to determine the target content as the associated search content if there is a target content in the candidate associated content and the search association degree corresponding to the target content meets the preset association condition.

[0379] In an optional embodiment, the acquisition unit 521 is also used to acquire a search record corresponding to the first search content within the historical time period, and the search record includes a record generated when the first search content is received within the historical time period; and acquire the candidate associated content corresponding to the first search content from the search record.

[0380] In an optional embodiment, the search relevance includes a relevance search rate;

[0381] The calculation unit 522 is further configured to calculate the frequency of occurrence of the i-th search content in the search record for the i-th search content in the candidate associated content, and obtain an associated search rate corresponding to the i-th search content, where i is a positive integer.

[0382] In an optional embodiment, the search relevance includes content relevance;

[0383] The calculation unit 522 is also used to calculate the i-th content correlation between the i-th search content and the first search content for the i-th search content in the candidate associated content, and the i-th content correlation is used to indicate the degree of correlation between the i-th search content and the first search content at the content topic and semantic levels.

[0384] In an optional embodiment, the calculation unit 522 is also used to calculate, for the i-th search content in the candidate associated content, the i-th edit distance and the i-th set similarity between the i-th search content and the first search content, the i-th edit distance referring to the number of single-character editing operations required to convert the first search content into the i-th search content, and the i-th set similarity referring to the ratio of the intersection and the union between the first search content and the i-th search content; obtain the content topic of the i-th search content and the content topic of the first search content; calculate the i-th topic correlation between the i-th search content and the first search content, the i-th topic correlation is used to indicate whether the content topic of the i-th search content and the content topic of the first search content meet a preset correlation condition; obtain the i-th content correlation based on the i-th edit distance, the i-th set similarity and the i-th topic correlation.

[0385] In an optional embodiment, the determining unit 523 is further configured to sort the candidate associated contents based on the search relevance, and determine the top j ranked contents among the candidate associated contents as the associated search contents, where j is a positive integer.

[0386] In an optional embodiment, the associated search content includes at least one sub-content;

[0387] The search module 530 is further used to obtain the search correlation between the at least one sub-content and the first search content, wherein the search correlation is used to describe the degree of correlation between the sub-content and the first search content; sort the at least one sub-content based on the search correlation to obtain a sorting result; search the first search content and the at least one sub-content to obtain a first search result corresponding to the first search content and a second search result corresponding to the at least one sub-content; and integrate the first search result and the second search result based on the sorting result to obtain the search result.

[0388] In summary, the content search device provided by the present application performs association analysis and historical tracing on the search content contained in the search request to obtain corresponding associated search content. The associated search content can reflect the user's satisfaction when receiving feedback on the same search content within a historical time period, and obtain the user's true intention when searching based on the first search content. On the basis of the original search content, the results obtained when searching for the associated search content are continuously added, and more accurate and comprehensive search results can be fed back to the user with a full understanding of the user's intention, thereby improving the accuracy of the user's search based on the search content. In addition, the frequency of search events in which users search again in a short period of time due to dissatisfaction with the search results is reduced to a certain extent, thereby improving the efficiency of user searches.

[0389] It should be noted that the content search device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the content search device provided in the above embodiment and the content search method embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0390] Figure 7 The block diagram of the structure of a computer device 700 provided by an exemplary embodiment of the present application is shown. The computer device 700 may be: a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3), an MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) player, a laptop computer or a desktop computer. The computer device 700 may also be called a user device, a portable terminal, a laptop terminal, a desktop terminal or other names.

[0391] Typically, the computer device 700 includes a processor 701 and a memory 702 .

[0392] The processor 701 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 701 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 701 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state.

[0393] In some embodiments, the processor 701 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 701 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0394] The memory 702 may include one or more computer-readable storage media, which may be non-transitory.

[0395] The memory 702 may also include high-speed random access memory, and non-volatile memory, such as one or more disk storage devices, flash memory storage devices.

[0396] In some embodiments, the non-transitory computer-readable storage medium in the memory 702 is used to store at least one instruction, and the at least one instruction is used to be executed by the processor 701 to implement the content search method provided by the method embodiment of the present application.

[0397] In some embodiments, the computer device 700 further includes some other components 703, and the type and quantity of the other components 703 can be selected based on the functional requirements of the computer device 700. Those skilled in the art will appreciate that Figure 7 The structure shown in the figure does not constitute a limitation on the computer device 700, and the computer device 700 may include more or less components than shown in the figure, or combine some components, or adopt a different arrangement of components.

[0398] Optionally, the computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a solid state drive (SSD), or an optical disk.

[0399] The random access memory may include a resistance random access memory (ReRAM) and a dynamic random access memory (DRAM). The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0400] An embodiment of the present application also provides a computer device, which includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement a content search method as described in any of the above embodiments of the present application.

[0401] An embodiment of the present application also provides a computer-readable storage medium, in which at least one instruction, at least one program, a code set or an instruction set is stored. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement a content search method as described in any of the above embodiments of the present application.

[0402] The embodiment of the present application also provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the content search method described in any of the above embodiments.

[0403] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0404] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A content search method, characterized in that: The method comprises: receiving a search request, wherein the search request includes a first search content, and the first search content is a search condition for data search; Acquire associated search content corresponding to the first search content, wherein the associated search content is search content whose search trigger interval with the first search content is within a preset time range within a historical time period, and the first search content and the associated search content meet a preset association condition; The first search content and the associated search content are searched to obtain search results.

2. The method according to claim 1, characterized in that Before acquiring the associated search content corresponding to the first search content, the method further includes: Acquire candidate associated content corresponding to the first search content, the candidate associated content being at least one search content whose search triggering interval duration with the first search content is within the preset duration range within the historical time period; Calculating the search relevance between the candidate related content and the first search content; In the case that there is a target content in the candidate related contents and the search relevance corresponding to the target content meets the preset relevance condition, the target content is determined as the related search content.

3. The method according to claim 2, characterized in that The acquiring candidate associated content corresponding to the first search content includes: Acquire a search record corresponding to the first search content within the historical time period, wherein the search record includes a record generated when the first search content is received within the historical time period; The candidate associated content corresponding to the first search content is acquired from the search record.

4. The method according to claim 3, characterized in that The search relevance includes a relevance search rate; The calculating the search relevance between the candidate related content and the first search content includes: For the i-th search content in the candidate related content, the frequency of occurrence of the i-th search content in the search record is calculated to obtain the related search rate corresponding to the i-th search content, where i is a positive integer.

5. The method according to claim 3, characterized in that: The search relevance includes content relevance; The calculating the search relevance between the candidate related content and the first search content includes: For the i-th search content in the candidate associated content, the i-th content correlation between the i-th search content and the first search content is calculated, and the i-th content correlation is used to indicate the degree of correlation between the i-th search content and the first search content at the content theme and semantic level.

6. The method according to claim 5, characterized in that The step of calculating, for the i-th search content in the candidate associated content, the i-th content correlation between the i-th search content and the first search content includes: For the i-th search content in the candidate associated content, calculating the i-th edit distance and the i-th set similarity between the i-th search content and the first search content, the i-th edit distance refers to the number of single-character edit operations required to convert the first search content into the i-th search content, and the i-th set similarity refers to the ratio of the intersection and union between the first search content and the i-th search content; Obtaining the content theme of the i-th search content and the content theme of the first search content; Calculating an i-th topic correlation between the i-th search content and the first search content, the i-th topic correlation being used to indicate whether a content topic of the i-th search content and a content topic of the first search content meet a preset correlation condition; The i-th content relevance is obtained based on the i-th edit distance, the i-th set similarity and the i-th topic relevance.

7. The method according to claim 2, characterized in that In the case where there is a target content in the candidate related content and the search relevance corresponding to the target content meets the preset relevance condition, determining the target content as the related search content includes: The candidate related contents are sorted based on the search relevance, and top j contents among the candidate related contents are determined as the related search contents, where j is a positive integer.

8. The method according to any one of claims 1 to 7, characterized in that: The associated search content includes at least one sub-content; The step of searching the first search content and the associated search content to obtain search results includes: Acquire a search relevance between the at least one sub-content and the first search content, wherein the search relevance is used to describe a relevance degree between the sub-content and the first search content; sorting the at least one sub-content based on the search relevance to obtain a sorting result; Searching the first search content and the at least one sub-content to obtain a first search result corresponding to the first search content and a second search result corresponding to each of the at least one sub-content; The first search result and the second search result are integrated based on the ranking result to obtain the search result.

9. A content search device, characterized in that: The device comprises: A receiving module, configured to receive a search request, wherein the search request includes a first search content, and the first search content is a search condition for data search; an acquisition module, configured to acquire associated search content corresponding to the first search content, wherein the associated search content is search content whose search trigger interval with the first search content is within a preset time range within a historical time period, and the first search content and the associated search content meet a preset association condition; The search module is used to search the first search content and the associated search content to obtain search results.

10. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one program, and the at least one program is loaded and executed by the processor to implement the content search method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The storage medium stores at least one program, and the at least one program is loaded and executed by the processor to implement the content search method according to any one of claims 1 to 8.

12. A computer program product, characterized in that The invention comprises a computer program, which implements the content search method according to any one of claims 1 to 8 when being executed by a processor.

Images