Content searching method and device and computer readable storage medium

Abstract

The embodiment of the invention discloses a content searching method and device and a computer readable storage medium. The method comprises the steps of obtaining search information input through a search function provided by a content interaction platform; searching in an interaction content database of the content interaction platform based on the search information, wherein interaction content generated in the content interaction platform is stored in the interaction content database; when the search fails, determining a target knowledge domain type corresponding to the search information, and searching target knowledge information matched with the search information in knowledge information under the target knowledge domain type of a knowledge database configured by the content interaction platform; and performing text generation on the target knowledge information based on a text generation model to obtain search reply information as a search result of the search information. Therefore, by constructing the knowledge database for the content interaction platform, the situation that the user inputs the search information and does not search the result can be avoided, the content search efficiency is improved, and then the content search experience is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a content search method, device and computer-readable storage medium. Background Art

[0002] With the rapid development of computer technology, a huge amount of content has been generated on the Internet. Users can search for the required content in the huge amount of content through search services. For example, for a content interaction platform, users can input search information through the search function provided by the content interaction platform, and then search for matching answers in the interactive content generated by the content interaction platform based on the search information.

[0003] In the process of research and practice of existing technologies, it was found that in the existing content search methods based on content interaction platforms, it is very easy for users to input search information but fail to find matching answers, resulting in low content search efficiency and a poor content search experience. Summary of the invention

[0004] The embodiments of the present application provide a content search method, device, and computer-readable storage medium, which can avoid the situation where a user enters search information but does not find any search results, thereby improving content search efficiency and further enhancing the content search experience.

[0005] The present application provides a content search method, including:

[0006] Obtain search information entered through the search function provided by the content interaction platform;

[0007] Searching an interactive content database of the content interaction platform based on the search information, wherein the interactive content database stores interactive content generated in the content interaction platform;

[0008] When the search fails, determining the target knowledge domain type corresponding to the search information, and searching for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured on the content interaction platform;

[0009] The target knowledge information is subjected to text generation based on a text generation model to obtain search reply information as a search result of the search information.

[0010] Accordingly, an embodiment of the present application provides a content search device, including:

[0011] An information acquisition unit, used to acquire search information input through a search function provided by the content interaction platform;

[0012] A first search unit, configured to search an interactive content database of the content interaction platform based on the search information, wherein the interactive content database stores interactive content generated in the content interaction platform;

[0013] A second search unit is used to determine the target knowledge field type corresponding to the search information when the search fails, and search for target knowledge information matching the search information in the knowledge information under the target knowledge field type in the knowledge database configured on the content interaction platform;

[0014] The search result generating unit is used to generate text for the target knowledge information based on a text generation model to obtain search reply information as the search result of the search information.

[0015] In one embodiment, the knowledge information in the knowledge database includes knowledge titles and knowledge contents, and the second search unit includes:

[0016] A knowledge title search subunit, used to obtain knowledge titles under the target knowledge domain type in the knowledge database configured by the content interaction platform;

[0017] Searching the target knowledge title matching the search information among the knowledge titles;

[0018] The target knowledge information determination subunit is used to determine the knowledge content corresponding to the target knowledge title as the target knowledge information associated with the search information.

[0019] In one embodiment, the number of the target knowledge information is at least two, and the search result generating unit includes:

[0020] An information generation subunit, configured to generate text generation instruction information according to the search information and the target knowledge information, wherein the text generation instruction information indicates to rewrite the target knowledge information based on the search information;

[0021] The text rewriting subunit is used to generate indication information and the search information based on the text, adopt a text generation model to perform text rewriting on the target knowledge information, and obtain search reply information as the search result of the search information.

[0022] In one embodiment, the text rewriting subunit is used to:

[0023] Based on the text generation instruction information and the search information, a text generation model is used to perform text rewriting on the target knowledge information to obtain search answer information corresponding to the target knowledge information;

[0024] Determining the degree of association between the target knowledge information and the search information, sorting the search answer information according to the degree of association, and obtaining sorted search answer information;

[0025] The sorted search reply information is determined as the search result of the search information.

[0026] In one embodiment, the content search device further includes:

[0027] An information content acquisition unit, used to acquire information content from multiple data sources;

[0028] A knowledge domain classification unit, used to classify the information content into knowledge domains to obtain candidate knowledge information of at least one knowledge domain;

[0029] A fact correction unit, used to perform fact correction on the candidate knowledge information using a fact detection model to obtain corrected knowledge information;

[0030] The knowledge base construction unit is used to construct a knowledge database corresponding to the content interaction platform based on the corrected knowledge information.

[0031] In one embodiment, the knowledge domain classification unit is used to:

[0032] Performing knowledge decomposition on the information content to obtain at least one knowledge information content corresponding to the information content;

[0033] Extracting knowledge titles and knowledge contents from the knowledge information content, and determining the extracted knowledge titles and corresponding knowledge contents as candidate knowledge information;

[0034] The candidate knowledge information is classified according to knowledge fields to obtain candidate knowledge information in at least one knowledge field.

[0035] In one embodiment, the fact detection model includes a timeliness detection model, a relevance model, and a content correction model, and the fact correction unit is used to:

[0036] Using a timeliness detection model to perform timeliness detection on the candidate knowledge information to obtain a timeliness relevance degree corresponding to the candidate knowledge information;

[0037] When the timeliness relevance degree meets a preset condition, searching is performed based on the candidate knowledge information to obtain search content corresponding to the candidate knowledge information;

[0038] Using a correlation model to determine the correlation between the search content and the candidate knowledge information, and filtering out target search content from the search content based on the correlation;

[0039] A content correction model is adopted to perform content correction on the candidate knowledge information based on the target search content and the candidate knowledge information to obtain corrected knowledge information.

[0040] In addition, an embodiment of the present application further provides a computer-readable storage medium, which stores a plurality of instructions, and the instructions are suitable for a processor to load to execute the steps in any content search method provided in the embodiment of the present application.

[0041] In addition, an embodiment of the present application further provides a computer device, including a processor and a memory, wherein the memory stores an application program, and the processor is used to run the application program in the memory to implement the content search method provided in the embodiment of the present application.

[0042] The embodiment of the present application also provides a computer program product or a computer program, wherein the computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the content search method provided in the embodiment of the present application.

[0043] The embodiment of the present application obtains the search information input through the search function provided by the content interaction platform; searches in the interactive content database of the content interaction platform based on the search information, and the interactive content generated in the content interaction platform is stored in the interactive content database; when the search fails, the target knowledge field type corresponding to the search information is determined, and the target knowledge information matching the search information is searched in the knowledge information under the target knowledge field type in the knowledge database configured by the content interaction platform; the text generation model is used to generate the target knowledge information, and the search reply information is obtained as the search result of the search information. In this way, by constructing a knowledge database with knowledge information of multiple knowledge fields for the content interaction platform, when the search fails in the interactive content database based on the input search information, the target knowledge field type corresponding to the search information can be determined, and the target knowledge information matching the search information is searched in the knowledge information under the target knowledge field type in the knowledge database configured by the content interaction platform, and then the search result of the search information is obtained according to the target knowledge information, thereby avoiding the situation where the user inputs the search information but fails to find the matching answer, improving the content search efficiency, and thus improving the content search experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] 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 those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 This is a schematic diagram of an implementation scenario of a content search method provided in an embodiment of the present application;

[0046] Figure 2 It is a flowchart of a content search method provided in an embodiment of the present application;

[0047] Figure 3 It is a schematic diagram of the overall process of a content search method provided by an embodiment of the present application;

[0048] Figure 4 It is a specific flow chart of a content search method provided by an embodiment of the present application;

[0049] Figure 5 is a schematic diagram of the structure of a content search device provided in an embodiment of the present application;

[0050] Figure 6 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0052] The embodiments of the present application provide a content search method, device and computer-readable storage medium. The content search device can be integrated in a computer device, which can be a server or a terminal.

[0053] Among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, network acceleration services (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms. Terminals may include but are not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, etc. Terminals and servers can be directly or indirectly connected via wired or wireless communications, and this application does not limit this.

[0054] See also Figure 1 , taking the case where the content search device is integrated into a computer device, Figure 1 A schematic diagram of an implementation scenario of the content search method provided in an embodiment of the present application, wherein the computer device can be a server or a terminal, and the computer device can obtain search information input through a search function provided by a content interaction platform; search in an interactive content database of the content interaction platform based on the search information, wherein the interactive content generated in the content interaction platform is stored in the interactive content database; when the search fails, determine the target knowledge domain type corresponding to the search information, and search for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured for the content interaction platform; generate text for the target knowledge information based on a text generation model, and obtain search reply information as the search result of the search information.

[0055] It should be noted that Figure 1 The implementation environment scenario diagram of the content search method shown is only an example. The implementation environment scenario of the content search method described in the embodiment of the present application is to more clearly illustrate the technical solution of the embodiment of the present application, and does not constitute a limitation on the technical solution provided by the embodiment of the present application. It is known to those skilled in the art that with the evolution of data processing and the emergence of new business scenarios, the technical solution provided by the present application is also applicable to similar technical problems.

[0056] The solutions provided by the embodiments of the present application are specifically described by the following embodiments. It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0057] This embodiment will be described from the perspective of a content search device. The content search device may be integrated into a computer device, which may be a server. This application does not limit this.

[0058] See also Figure 2 , Figure 2: is a flow chart of a content search method provided in an embodiment of the present application. The content search method includes:

[0059] In step 101, search information input through a search function provided by a content interaction platform is obtained.

[0060] Among them, the content interaction platform can be a platform for content interaction. For example, users can create notes, copywriting, videos and other content and publish them in the content interaction platform. Users can also watch, comment, save, share and other interactive behaviors based on the content published by other users in the content interaction platform. The search function can be a function for content search provided in the content interaction platform. For example, a search box can be provided in the page of the content interaction platform, and users can enter the information to be searched through the search box. The search information can be information that needs to be searched, for example, it can be a question that needs to be searched entered by the user. For example, the search information can be "Which variety of apples is the most delicious" or "XX's travel guide". Optionally, the user can enter the search information in the form of text or in the form of voice.

[0061] In step 102, a search is performed in an interactive content database of a content interaction platform based on the search information.

[0062] The interactive content database may be a database of a content interaction platform, and may be used to store interactive content generated in the content interaction platform. The interactive content may be content generated in the content interaction platform for content interaction, such as notes, videos, and texts. The interactive content generated in the content interaction platform may include content created by users on the content interaction platform, or content uploaded by users to the content interaction platform.

[0063] Optionally, when target interactive content matching the search information is searched in the interactive content database of the content interactive platform based on the search information, the target interactive content can be fed back to the user as the search result of the search information. The target interactive content can be interactive content matching the search information.

[0064] In step 103, when the search fails, the target knowledge domain type corresponding to the search information is determined, and the target knowledge information matching the search information is searched in the knowledge information under the target knowledge domain type in the knowledge database configured in the content interaction platform.

[0065] Among them, search failure may refer to the situation where no content matching the search information is found in the interactive content database. The target knowledge domain type may be the knowledge domain type corresponding to the search information. The knowledge domain type may be the type of knowledge domain to which the search information belongs. The knowledge domain may be a domain determined according to the knowledge requirements. For example, it may include knowledge domains such as medicine, education, history, mathematics, and science. The division of specific domains may be determined according to actual conditions. The knowledge database may be a database configured in the content interaction platform that stores knowledge in various knowledge domains. The knowledge information may be a storage form of each knowledge stored in the knowledge database. For example, the knowledge may be described in the form of text, voice, picture, or video. For example, assuming that a piece of knowledge is content describing the formation principle of cumulus clouds, the knowledge information may be text or picture describing the formation principle of cumulus clouds, or audio or video describing the formation principle of cumulus clouds, or information describing the formation principle of cumulus clouds in a mixed form of text and picture, etc. The target knowledge information may be knowledge information that matches the search information. For example, assuming that the user inputs the search information "Why do cumulus clouds appear in the sky", it may be determined that the knowledge domain type of the search information is a scientific type, that is, the target knowledge domain type is a scientific type. Thus, the target knowledge information that matches the search information may be searched for in the knowledge information under the scientific type of the knowledge database configured on the content interaction platform. For example, the knowledge information that describes the formation principle of cumulus clouds may be determined as the target knowledge information that matches the search information "Why do cumulus clouds appear in the sky".

[0066] Among them, there may be multiple ways to search for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured for the content interaction platform. For example, the knowledge information stored in the knowledge database may include knowledge titles and knowledge content, so that the knowledge titles under the target knowledge domain type in the knowledge database configured for the content interaction platform can be obtained, and in the knowledge titles, the target knowledge titles matching the search information are searched, and the knowledge content corresponding to the target knowledge title is determined as the target knowledge information associated with the search information.

[0067] The knowledge title may be the title of the knowledge information, and the knowledge content may be the content of the knowledge information. For example, assuming that a piece of knowledge information is information describing the formation principle of cumulus clouds, the knowledge title of the knowledge information may be a description such as "formation principle of cumulus clouds", and the knowledge content of the knowledge information may be content describing the formation principle of cumulus clouds, for example, content in the form of text, video, image, etc. that can describe the formation principle of cumulus clouds. In a knowledge database, each piece of knowledge information may be divided into a knowledge title and corresponding knowledge content for storage, so that target knowledge information matching the search information may be found in the knowledge database according to the knowledge title of the knowledge information, and then feedback of the search results may be performed according to the knowledge content in the target knowledge information. The target knowledge title may be a knowledge title matching the search information.

[0068] In one embodiment, there may be multiple methods for constructing a knowledge database of a content interaction platform. For example, information content may be obtained from multiple data sources, the information content may be classified into knowledge fields to obtain candidate knowledge information of at least one knowledge field, a fact detection model may be used to perform fact correction on the candidate knowledge information to obtain corrected knowledge information, and a knowledge database corresponding to the content interaction platform may be constructed based on the corrected knowledge information.

[0069] The information content may be content obtained from multiple data sources, and the content may contain information of multiple knowledge. The candidate knowledge information may be information containing knowledge under each knowledge field obtained by classifying the information in the information content by knowledge field, and the fact detection model may be a model for detecting whether the knowledge contained in the candidate knowledge information is correct (conforms to the facts), so as to ensure the factual accuracy of the knowledge database, provide the accuracy of content search, and thus improve the search experience.

[0070] There are many ways to classify information content into knowledge fields and obtain candidate knowledge information in at least one knowledge field. For example, the information content can be decomposed into knowledge to obtain at least one knowledge information content corresponding to the information content, and knowledge titles and knowledge contents can be extracted from the knowledge information content. The extracted knowledge titles and corresponding knowledge contents can be determined as candidate knowledge information, and the candidate knowledge information can be classified into knowledge fields to obtain candidate knowledge information under at least one knowledge field.

[0071] The knowledge information content may be the content corresponding to each knowledge obtained by decomposing the information content based on the knowledge point dimension. For example, if the information content obtained is the promotional content about XX enterprise, the information content may be decomposed based on the knowledge dimension, for example, it may be decomposed into the information content corresponding to the knowledge points such as the founder of XX enterprise, the location of XX enterprise, the main business of XX enterprise and the establishment time of XX enterprise.

[0072] In one embodiment, please refer to Figure 3 , Figure 3 It is an overall flow chart of a content search method provided in an embodiment of the present application, which can obtain information content from multiple data sources and decompose the information content into at least one knowledge information content, then extract knowledge titles and knowledge content from the knowledge information content, merge the extracted knowledge titles and corresponding knowledge content into candidate knowledge information, classify the candidate knowledge information into knowledge fields, and obtain candidate knowledge information of at least one knowledge field, then, use a fact detection model to fact-correct the candidate knowledge information to obtain the corrected knowledge information, so that a knowledge database corresponding to the content interaction platform can be constructed based on the corrected knowledge information.

[0073] After classifying the information content into knowledge domains and obtaining candidate knowledge information in at least one knowledge domain, the fact detection model can be used to fact-correct the candidate knowledge information to obtain the corrected knowledge information. There are many ways to fact-correct the candidate knowledge information using the fact detection model. For example, please refer to Figure 4 , Figure 4 It is a specific flow chart of a content search method provided in an embodiment of the present application. The fact detection model may include a timeliness detection model, a relevance model and a content correction model. Thus, the timeliness detection model may be used to perform timeliness detection on candidate knowledge information to obtain the timeliness relevance degree corresponding to the candidate knowledge information. When the timeliness relevance degree meets the preset conditions, a search is performed based on the candidate knowledge information to obtain the search content corresponding to the candidate knowledge information. The relevance model is used to determine the relevance between the search content and the candidate knowledge information. Based on the relevance, a target search content is screened out from the search content. The content correction model is used to perform content correction on the candidate knowledge information based on the target search content and the candidate knowledge information to obtain the corrected knowledge information.

[0074] Among them, the timeliness detection model can be a model for detecting the timeliness of candidate knowledge information, and the timeliness relevance degree can be information for measuring the timeliness degree of candidate knowledge information. Timeliness can refer to the fact that the same thing has great differences in nature at different times. This difference is also called timeliness. When the timeliness relevance degree is high, the same knowledge will have great differences at different times. For example, since the capital of Brazil in 1949 was Rio de Janeiro and the current capital of Brazil is Brasilia, the knowledge of the capital of Brazil is strongly related to timeliness. The preset condition can be a condition indicating that the timeliness relevance degree meets the timeliness relevance degree. For example, the preset condition can include a threshold value of the timeliness relevance degree. When the timeliness relevance degree of the candidate knowledge information is greater than the threshold value, the candidate knowledge information can be considered to be knowledge with strong timeliness relevance. Therefore, subsequent fact detection and correction steps are required to improve the factual accuracy of the knowledge in the knowledge database, thereby improving the accuracy of the knowledge database and improving the content search quality. When the timeliness relevance of the candidate knowledge information is not greater than the threshold, the candidate knowledge information can be considered as knowledge with weak timeliness relevance, and in this case, subsequent steps such as fact correction can be omitted. The search content can be content related to the candidate knowledge information searched through a search engine or other search plug-in. For example, when the candidate knowledge information is the capital of Brazil, the search content can be content searched for information containing the capital of Brazil. Optionally, the search content can be obtained by searching according to the knowledge title of the candidate knowledge information. The relevance model may be a model for determining the degree of correlation between candidate knowledge information and search content, the target search content may be the search content with the highest correlation with the candidate knowledge information in the search content, and the content correction model may be a model for performing content correction on the candidate knowledge information. Optionally, the content correction model may adopt a text processing model such as a generative pre-trained Transformer model (GPT for short) as a basic model structure. The content correction model may determine whether the candidate knowledge information is consistent with the facts based on the input target search content and the information described by the candidate knowledge information, so that when the candidate knowledge information is not consistent with the facts, the candidate knowledge information may be corrected according to the target search content, and when the candidate knowledge information is consistent with the facts, the content of the candidate knowledge information may be retained.

[0075] Among them, a content correction model is adopted to perform content correction on the candidate knowledge information based on the target search content and the candidate knowledge information. There are many ways to obtain the corrected knowledge information. For example, the target search content and the candidate knowledge information can be input into the content correction model, and the content correction model is used to predict whether the candidate knowledge information is correct. When the candidate knowledge information is incorrect, the candidate knowledge information can be corrected according to the target search content. Optionally, the target search content can also be integrated into the candidate knowledge information to increase the content richness of the candidate knowledge information.

[0076] In step 104, text generation is performed on the target knowledge information based on the text generation model to obtain search response information as the search result of the search information.

[0077] The text generation model may be a model for text generation, and the text generation model may be used to generate text as search feedback for the search information based on the target knowledge information. The search reply information may be information for search reply feedback for the search information, and the search reply information may include reply information for the search information and the target knowledge information. Optionally, the text generation model may use a text processing model such as a GPT model as a basic model structure.

[0078] For example, when the search information is "Why do cumulus clouds appear in the sky", the target knowledge information is "Cumulus clouds are a type of cloud, about 5,500 meters high, with very small, white and shadowless clouds. They are composed of white wavy, scale-like or spherical small cloud pieces or cloud sheets or cloud layers, often arranged in rows or groups, much like the small ripples caused by a breeze blowing across the water. White with no shadows, it has a silky luster. The cloud body is very thin and can pass through the sun and moonlight. It is white without dark shapes. At night it is gray-black and is almost entirely composed of ice crystals." According to the target knowledge information, corresponding reply information can be generated. For example, "Aren't the cumulus clouds floating in the sky beautiful? Did you know that the appearance of cumulus clouds in the sky indicates that there may be unstable weather?"

[0079] There are many ways to generate text for target knowledge information based on a text generation model and obtain search reply information as the search result of the search information. For example, text generation instruction information can be generated based on the search information and the target knowledge information, and the text generation instruction information and search information can be rewritten using a text generation model to obtain search reply information as the search result of the search information.

[0080] The text generation instruction information may be information indicating that the target knowledge information is to be rewritten based on the search information. For example, it may be information indicating that a reply to the search information is to be generated in the form of a dialogue according to the content of the target knowledge information. The specific content of the text generation instruction information may be set according to specific needs, and this application does not limit this. Optionally, when the text generation model adopts a text processing model such as a GPT model as the basic model structure, the text generation instruction information may be a text prompt of the text generation model.

[0081] In this way, by searching for target knowledge information matching the search information in the knowledge information stored in the knowledge database configured in the content interaction platform, a text generation model can be used to generate search results matching the search information based on the target knowledge information, thereby avoiding the situation where users enter search information in the content interaction platform but fail to find a matching answer. At the same time, the search results are fed back in the form of reply information, which can facilitate users to view the information in the search results. At the same time, it can improve the user's experience of viewing the search results, thereby improving the content search efficiency and effectively improving the user's experience of content search.

[0082] Among them, there can be multiple ways to generate indication information and search information based on text, rewrite the target knowledge information with a text generation model, and obtain search reply information as the search result of the search information. For example, the number of target knowledge information can be at least two, generate indication information and search information based on text, rewrite the target knowledge information with a text generation model, obtain search reply information corresponding to the target knowledge information, determine the degree of correlation between the target knowledge information and the search information, sort the search reply information according to the degree of correlation, obtain sorted search reply information, and determine the sorted search reply information as the search result of the search information.

[0083] The correlation degree may be information measuring the degree of matching between the target knowledge information and the search information, and the sorted search reply information may be search reply information obtained by sorting the search reply information corresponding to the target knowledge information according to the correlation degree. Since the correlation degree between each target knowledge information and the search information may be different, the search reply information corresponding to the target knowledge information may be sorted according to the correlation degree, so that the search reply information corresponding to the target knowledge information with a greater correlation degree is fed back to the user first, so that the user can obtain the required search results more quickly, thereby improving the content search efficiency.

[0084] As can be seen from the above, the embodiment of the present application obtains the search information input through the search function provided by the content interaction platform; searches in the interactive content database of the content interaction platform based on the search information, and the interactive content generated in the content interaction platform is stored in the interactive content database; when the search fails, the target knowledge field type corresponding to the search information is determined, and the target knowledge information matching the search information is searched in the knowledge information under the target knowledge field type in the knowledge database configured by the content interaction platform; the text generation model is used to generate the target knowledge information, and the search reply information is obtained as the search result of the search information. In this way, by constructing a knowledge database with knowledge information of multiple knowledge fields for the content interaction platform, when the search fails in the interactive content database based on the input search information, the target knowledge field type corresponding to the search information can be determined, and the target knowledge information matching the search information can be searched in the knowledge information under the target knowledge field type in the knowledge database configured by the content interaction platform, and then the search result of the search information can be obtained according to the target knowledge information, thereby avoiding the situation where the user inputs the search information but fails to find a matching answer, improving the content search efficiency, and thus improving the content search experience.

[0085] In order to better implement the above method, an embodiment of the present invention further provides a content search device, which can be integrated in a computer device, and the computer device can be a server.

[0086] For example, Figure 5 As shown, it is a schematic diagram of the structure of the content search device provided in an embodiment of the present application. The content search device may include an information acquisition unit 201, a first search unit 202, a second search unit 203 and a search result generation unit 204, as follows:

[0087] The information acquisition unit 201 is used to acquire search information input through the search function provided by the content interaction platform;

[0088] A first search unit 202, configured to search in an interactive content database of the content interaction platform based on the search information, wherein the interactive content generated in the content interaction platform is stored in the interactive content database;

[0089] The second search unit 203 is used to determine the target knowledge domain type corresponding to the search information when the search fails, and search for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured in the content interaction platform;

[0090] The search result generating unit 204 is used to generate text for the target knowledge information based on the text generation model, and obtain search reply information as the search result of the search information.

[0091] In one embodiment, the knowledge information in the knowledge database includes knowledge title and knowledge content, and the second search unit 203 includes:

[0092] A knowledge title search subunit is used to obtain knowledge titles under a target knowledge domain type in a knowledge database configured in the content interaction platform;

[0093] In the knowledge titles, find the target knowledge titles that match the search information;

[0094] The target knowledge information determination subunit is used to determine the knowledge content corresponding to the target knowledge title as the target knowledge information associated with the search information.

[0095] In one embodiment, the number of target knowledge information is at least two, and the search result generating unit 204 includes:

[0096] An information generation subunit, used to generate text generation instruction information according to the search information and the target knowledge information, wherein the text generation instruction information indicates to rewrite the target knowledge information based on the search information;

[0097] The text rewriting subunit is used to generate indication information and search information based on the text, adopt the text generation model to rewrite the target knowledge information, and obtain the search response information as the search result of the search information.

[0098] In one embodiment, the text rewriting subunit is used to:

[0099] Based on the text generation instruction information and search information, the text generation model is used to rewrite the target knowledge information to obtain the search response information corresponding to the target knowledge information;

[0100] Determine the correlation degree between the target knowledge information and the search information, sort the search response information according to the correlation degree, and obtain sorted search response information;

[0101] The sorted search reply information is determined as the search result of the search information.

[0102] In one embodiment, the content search device further includes:

[0103] An information content acquisition unit, used to acquire information content from multiple data sources;

[0104] A knowledge domain classification unit, used to classify the information content into knowledge domains to obtain candidate knowledge information of at least one knowledge domain;

[0105] A fact correction unit, used for performing fact correction on candidate knowledge information by using a fact detection model to obtain corrected knowledge information;

[0106] The knowledge base construction unit is used to construct a knowledge database corresponding to the content interaction platform based on the corrected knowledge information.

[0107] In one embodiment, the knowledge domain classification unit is used to:

[0108] Performing knowledge decomposition on the information content to obtain at least one knowledge information content corresponding to the information content;

[0109] Extracting knowledge titles and knowledge contents from the knowledge information content, and determining the extracted knowledge titles and corresponding knowledge contents as candidate knowledge information;

[0110] The candidate knowledge information is classified according to knowledge fields to obtain candidate knowledge information in at least one knowledge field.

[0111] In one embodiment, the fact detection model includes a timeliness detection model, a relevance model, and a content correction model, and the fact correction unit is used to:

[0112] The timeliness detection model is used to detect the timeliness of candidate knowledge information, and the timeliness relevance degree corresponding to the candidate knowledge information is obtained;

[0113] When the timeliness relevance degree meets the preset conditions, a search is performed based on the candidate knowledge information to obtain the search content corresponding to the candidate knowledge information;

[0114] A relevance model is used to determine the relevance between the search content and the candidate knowledge information, and the target search content is screened out from the search content based on the relevance;

[0115] A content correction model is adopted to correct the candidate knowledge information based on the target search content and the candidate knowledge information to obtain the corrected knowledge information.

[0116] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0117] As can be seen from the above, the embodiment of the present application obtains the search information input through the search function provided by the content interaction platform through the information acquisition unit 201; the first search unit 202 searches in the interactive content database of the content interaction platform based on the search information, and the interactive content database stores the interactive content generated in the content interaction platform; the second search unit 203 determines the target knowledge field type corresponding to the search information when the search fails, and searches for the target knowledge information matching the search information in the knowledge information under the target knowledge field type of the knowledge database configured in the content interaction platform; the search result generation unit 204 generates text for the target knowledge information based on the text generation model, and obtains the search reply information as the search result of the search information. In this way, by constructing a knowledge database with knowledge information of multiple knowledge fields for the content interaction platform, when the search fails in the interactive content database based on the input search information, the target knowledge field type corresponding to the search information can be determined, and the target knowledge information matching the search information can be searched in the knowledge information under the target knowledge field type of the knowledge database configured in the content interaction platform, and then the search result of the search information can be obtained according to the target knowledge information, thereby avoiding the situation where the user inputs the search information but fails to find a matching answer, improving the content search efficiency, and thus improving the content search experience.

[0118] The present application also provides a computer device, such as Figure 6 As shown, it shows a schematic diagram of the structure of a computer device involved in an embodiment of the present application, and the computer device may be a server, specifically:

[0119] The computer device may include one or more processing core processors 301, one or more computer readable storage media memories 302, a power supply 303, an input unit 304 and other components. Those skilled in the art will appreciate that Figure 6 The computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. Among them:

[0120] The processor 301 is the control center of the computer device. It uses various interfaces and lines to connect various parts of the entire computer device. It executes various functions of the computer device and processes data by running or executing software programs and / or modules stored in the memory 302, and calling data stored in the memory 302. Optionally, the processor 301 may include one or more processing cores; preferably, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 301.

[0121] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and data processing by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0122] The computer device also includes a power supply 303 for supplying power to various components. Preferably, the power supply 303 can be logically connected to the processor 301 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 303 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0123] The computer device may further include an input unit 304, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0124] Although not shown, the computer device may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 301 in the computer device will load the executable files corresponding to the processes of one or more application programs into the memory 302 according to the following instructions, and the processor 301 will run the application programs stored in the memory 302, thereby realizing various functions, as follows:

[0125] Acquire search information input through a search function provided by a content interaction platform; search in an interactive content database of the content interaction platform based on the search information, wherein the interactive content generated in the content interaction platform is stored in the interactive content database; when the search fails, determine the target knowledge domain type corresponding to the search information, and search for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured on the content interaction platform; generate text for the target knowledge information based on a text generation model, and obtain search reply information as the search result of the search information.

[0126] The specific implementation of each of the above operations can be found in the previous embodiments, which will not be described in detail here. It should be noted that the computer device provided in the embodiment of this application and the content search method in the above embodiment belong to the same concept, and its specific implementation process is detailed in the above method embodiment, which will not be described in detail here.

[0127] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0128] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any of the content search methods provided in the embodiments of the present application. For example, the instructions can execute the following steps:

[0129] Acquire search information input through a search function provided by a content interaction platform; search in an interactive content database of the content interaction platform based on the search information, wherein the interactive content generated in the content interaction platform is stored in the interactive content database; when the search fails, determine the target knowledge domain type corresponding to the search information, and search for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured on the content interaction platform; generate text for the target knowledge information based on a text generation model, and obtain search reply information as the search result of the search information.

[0130] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0131] Since the instructions stored in the computer-readable storage medium can execute the steps in any content search method provided in the embodiments of the present application, the beneficial effects that can be achieved by any content search method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0132] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, and 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 the methods provided in the various optional implementations provided in the above embodiments.

[0133] The above is a detailed introduction to a content search method, device and computer-readable storage medium provided in an embodiment of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A content search method, characterized in that: include: Obtain search information entered through the search function provided by the content interaction platform; Searching an interactive content database of the content interaction platform based on the search information, wherein the interactive content database stores interactive content generated in the content interaction platform; When the search fails, determining the target knowledge domain type corresponding to the search information, and searching for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured on the content interaction platform; The target knowledge information is subjected to text generation based on a text generation model to obtain search reply information as a search result of the search information.

2. The content search method according to claim 1, characterized in that: The knowledge information in the knowledge database includes knowledge titles and knowledge contents, and searching for target knowledge information matching the search information in the knowledge information under the target knowledge domain type in the knowledge database configured on the content interaction platform includes: Acquire the knowledge title under the target knowledge domain type in the knowledge database configured by the content interaction platform; Searching the target knowledge title matching the search information among the knowledge titles; The knowledge content corresponding to the target knowledge title is determined as the target knowledge information associated with the search information.

3. The content search method according to claim 1, characterized in that: The step of generating text for the target knowledge information based on the text generation model to obtain search answer information as the search result of the search information includes: generating text generation instruction information according to the search information and the target knowledge information, wherein the text generation instruction information indicates to rewrite the target knowledge information based on the search information; Based on the text generation instruction information and the search information, a text generation model is used to perform text rewriting on the target knowledge information to obtain search response information as a search result of the search information.

4. The content search method according to claim 3, characterized in that: The number of the target knowledge information is at least two, and the text generation instruction information and the search information are based on which the text generation model is used to rewrite the target knowledge information to obtain search answer information as the search result of the search information, including: Based on the text generation instruction information and the search information, a text generation model is used to perform text rewriting on the target knowledge information to obtain search answer information corresponding to the target knowledge information; Determining the degree of association between the target knowledge information and the search information, sorting the search answer information according to the degree of association, and obtaining sorted search answer information; The sorted search reply information is determined as the search result of the search information.

5. The content search method according to any one of claims 1 to 4, characterized in that: Before searching the knowledge information under the target knowledge domain type in the knowledge database configured in the content interaction platform for target knowledge information matching the search information, the method further includes: Obtain information content from multiple data sources; Classifying the information content by knowledge domain to obtain candidate knowledge information of at least one knowledge domain; Using a fact detection model to fact-correct the candidate knowledge information to obtain corrected knowledge information; A knowledge database corresponding to the content interaction platform is constructed based on the corrected knowledge information.

6. The content search method according to claim 5, characterized in that: The step of classifying the information content by knowledge domain to obtain candidate knowledge information of at least one knowledge domain includes: Performing knowledge decomposition on the information content to obtain at least one knowledge information content corresponding to the information content; Extracting knowledge titles and knowledge contents from the knowledge information content, and determining the extracted knowledge titles and corresponding knowledge contents as candidate knowledge information; The candidate knowledge information is classified according to knowledge fields to obtain candidate knowledge information in at least one knowledge field.

7. The content search method according to claim 5, characterized in that: The fact detection model includes a timeliness detection model, a relevance model, and a content correction model. The fact detection model is used to perform fact correction on the candidate knowledge information to obtain the corrected knowledge information, including: Using a timeliness detection model to perform timeliness detection on the candidate knowledge information to obtain a timeliness relevance degree corresponding to the candidate knowledge information; When the timeliness relevance degree meets a preset condition, searching is performed based on the candidate knowledge information to obtain search content corresponding to the candidate knowledge information; Using a correlation model to determine the correlation between the search content and the candidate knowledge information, and filtering out target search content from the search content based on the correlation; A content correction model is adopted to perform content correction on the candidate knowledge information based on the target search content and the candidate knowledge information to obtain corrected knowledge information.

8. A content search device, characterized in that: include: An information acquisition unit, used to acquire search information input through a search function provided by the content interaction platform; A first search unit, configured to search an interactive content database of the content interaction platform based on the search information, wherein the interactive content database stores interactive content generated in the content interaction platform; A second search unit is used to determine the target knowledge field type corresponding to the search information when the search fails, and search for target knowledge information matching the search information in the knowledge information under the target knowledge field type in the knowledge database configured on the content interaction platform; The search result generating unit is used to generate text for the target knowledge information based on a text generation model to obtain search reply information as the search result of the search information.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the content search method according to any one of claims 1 to 7.

10. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the content search method according to any one of claims 1 to 7 when executing the computer program.

Images