Service search method, device, computer equipment and storage medium

Abstract

The present application relates to a service search method, device, computer equipment and storage medium. The method obtains a service search request by obtaining a service search request; reading the original service search result corresponding to the query term; identifying the service knowledge graph association relationship between the query term and the original service search result; extracting data in the original service search result that has a service knowledge graph upper-lower relationship with the query term, and obtaining a target service search result. The service search method of the present application obtains a service search request by obtaining the original service search result by identifying the service knowledge graph association relationship between the query term and the original service search result; and extracting data in the original service search result that has a service knowledge graph upper-lower relationship with the query term, and obtaining the final target service search result based on these data. The obtained target service search result can better reflect the association relationship with the query term, and can effectively improve the search accuracy of the service search process.

Description

Technical Field

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

[0002] With the development of computer technology and search technology, vertical search technology has emerged. Vertical search is a search service provided by vertical search engines. Vertical search engines are professional search engines for a certain industry. They are a subdivision and extension of search engines. Vertical search is based on specific user's specific search request, and after deep mining and integration of a certain type of special information in the website (page) library, the results are returned to the user in some form. Vertical search is a new search engine service model proposed in response to the large amount of information, inaccurate query, and insufficient depth of search results obtained by general search engines. It provides information and related services with specific purposes for a specific field, a specific group of people, or a specific need. Service search is a type of vertical search, which can directly display services that meet user query requests to users. For example, when searching for a nanny, a nanny service menu can be directly provided to users through service search.

[0003] For service search, it is necessary to determine the semantic relevance between the query request and the query result during the search process. Currently, the semantic relevance between the query request and the query result is generally determined by a semantic model, but the semantic model can only identify the superficial relevance of the text, resulting in low accuracy of service search. Summary of the invention

[0004] Based on this, it is necessary to provide a service search method, device, computer equipment and storage medium that can improve the accuracy of service search in response to the above technical problems.

[0005] A service search method, the method comprising:

[0006] Obtaining a service search request, wherein the service search request carries a query term;

[0007] Reading the original service search result corresponding to the query term;

[0008] Identify the service knowledge graph association relationship between the query term and the original service search result;

[0009] Extract data in the original service search results that have a service knowledge graph hyponym relationship with the query term to obtain a target service search result.

[0010] A service search device, the device comprising:

[0011] A request acquisition module, used to acquire a service search request, wherein the service search request carries a query term;

[0012] An original result search module, used to read the original service search results corresponding to the query term;

[0013] An association relationship identification module, used to identify the association relationship between the query term and the service knowledge graph of the original service search result;

[0014] The target result identification module is used to extract data in the original service search results that have a service knowledge graph hyponym relationship with the query term to obtain the target service search results.

[0015] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0016] Obtaining a service search request, wherein the service search request carries a query term;

[0017] Reading the original service search result corresponding to the query term;

[0018] Identify the service knowledge graph association relationship between the query term and the original service search result;

[0019] Extract data in the original service search results that have a service knowledge graph hyponym relationship with the query term to obtain a target service search result.

[0020] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0021] Obtaining a service search request, wherein the service search request carries a query term;

[0022] Reading the original service search result corresponding to the query term;

[0023] Identify the service knowledge graph association relationship between the query term and the original service search result;

[0024] Extract data in the original service search results that have a service knowledge graph hyponym relationship with the query term to obtain a target service search result.

[0025] The above-mentioned service search method, apparatus, computer equipment and storage medium obtain a service search request, which carries a query term; reads the original service search result corresponding to the query term; identifies the service knowledge graph association relationship between the query term and the original service search result; and extracts data in the original service search result that has a service knowledge graph superior-female relationship with the query term, thereby obtaining a target service search result. The service search method of the present application obtains the target service search result based on these data by identifying the service knowledge graph association relationship between the query term and the original service search result after obtaining the original service search result, and extracting data in the original service search result that has a service knowledge graph superior-female relationship with the query term. The target service search result obtained can better reflect the association relationship with the query term, and can effectively improve the search accuracy of the service search process. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 An application environment diagram of a service search method in an embodiment;

[0027] Figure 2 A schematic diagram of a flow chart of a service search method in an embodiment;

[0028] Figure 3 A flowchart of the steps of obtaining original service search results in one embodiment;

[0029] Figure 4 A flowchart of a step of obtaining an association relationship between a query term and an original service search result in a preset service knowledge graph in one embodiment;

[0030] Figure 5 A schematic diagram of a process of constructing a first entity node set in an embodiment;

[0031] Figure 6 A schematic diagram of a flow chart of a step of constructing a first entity node set in another embodiment;

[0032] Figure 7 A schematic diagram of a process for constructing a preset service knowledge graph in an embodiment;

[0033] Figure 8 A schematic diagram of a partial structure of a preset service knowledge graph in an embodiment;

[0034] Fig. 9 A schematic diagram of a hierarchical structure for inferring correlation based on the hyponymy of multiple nodes in a graph in an embodiment;

[0035] Fig.10 is a structural block diagram of a vertical search device in one embodiment;

[0036] Fig.11FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0038] Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.

[0039] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. The basic technologies of artificial intelligence generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0040] The solution provided in the embodiment of the present application relates to the field of natural language processing (NLP) of artificial intelligence, and natural language processing is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that can realize effective communication between people and computers using natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Therefore, research in this field will involve natural language, that is, the language used by people in daily life, so it is closely related to the study of linguistics. Natural language processing technology generally includes technologies such as text processing, semantic understanding, machine translation, robot question and answer, knowledge graph, etc. The solution of this application is specifically illustrated by the following embodiments:

[0041] The service search method provided by this application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The terminal 102 can send a service search request to the server 104 to perform a related service search through the server 104 and obtain the final search results. The server 104 obtains the service search request submitted by the terminal 102; obtains the original service search result corresponding to the query term of the service search request; identifies the association relationship between the query term and the original service search result in the preset service knowledge graph; based on the association relationship, determines the target service search result in the original service search result. Then the final service search result is returned to the terminal 102. Among them, the terminal 102 can be but not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices, and the server 104 can be implemented with an independent server or a server cluster composed of multiple servers. In another embodiment, it is not necessary that the service search method of the present application can also be applied to the terminal, and the user can directly execute the method on the terminal side. In one embodiment, multiple servers can be composed of a blockchain, and the server is a node on the blockchain.

[0042] Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0043] The underlying blockchain platform can include processing modules such as user management, basic services, smart contracts, and operation monitoring. Among them, the user management module is responsible for the identity information management of all blockchain participants, including maintaining public and private key generation (account management), key management, and the maintenance of the correspondence between the user's real identity and the blockchain address (authority management), etc., and, under authorization, supervises and audits the transactions of certain real identities and provides risk control rule configuration (risk control audit); the basic service module is deployed on all blockchain node devices to verify the validity of business requests, and records valid requests to storage after consensus is reached. For a new business request, the basic service first adapts the interface for parsing and authentication (interface adaptation), and then encrypts the business information through the consensus algorithm (consensus management). The smart contract module is responsible for the registration and issuance of contracts, as well as contract triggering and contract execution. Developers can define the contract logic in a programming language and publish it to the blockchain (contract registration). According to the logic of the contract terms, the key or other events are called to trigger the execution and complete the contract logic. It also provides the function of contract upgrade and cancellation. The operation monitoring module is mainly responsible for the deployment, configuration modification, contract setting, cloud adaptation and real-time status visualization output of the product during the product release process, such as alarm, network status monitoring, node equipment health status monitoring, etc.

[0044] The platform product service layer provides the basic capabilities and implementation framework of typical applications. Developers can superimpose business features based on these basic capabilities to complete the blockchain implementation of business logic. The application service layer provides application services based on blockchain solutions for business participants to use.

[0045] In one embodiment, Figure 2 As shown, a service search method is provided, which is applied to Figure 1 Taking the server 104 in the example as an example, the following steps are included:

[0046] Step 201: the server 104 obtains a service search request, which carries a query term.

[0047] Among them, the service search method of the present application is specifically a vertical search method. The search service of vertical search is information and related services with specific purposes provided for a specific field, a specific group of people or a specific demand, and service search is a further vertical search method, which can directly display services that meet the user's query request to the user. The service search request is generated by the terminal 102 and is used to request a corresponding service from the server 104. The service search request specifically includes the query terms entered by the user, and the server 104 can identify the existing services searched by the service search request based on the query terms. Among them, the query term is query, which specifically refers to the content (query condition) entered by the user in the search engine, generally input keywords. For vertical search engines, categories can also be entered, such as "movies" and "TV series" can be selected in the Youku website. In e-commerce websites, various product brands, models, styles, prices, etc. are also common query conditions.

[0048] Specifically, when the operator of the terminal 102 needs to request a certain service, the information related to the service is obtained from the server 104, and a service search request can be generated through the terminal 102, and then the service search work is started by submitting the service search request to the server 104. Specifically, the service search method of the present application can be applied to an instant messaging application, which integrates a search function, and the search function can complete the search of user circles, articles, public accounts, novels, music, emoticons, services and other functions according to the query terms entered by the user. In addition, the user can also limit the search function, specify the search process as searching for a certain service, and then the user generates a service search request on the terminal side of the instant messaging application by entering the query terms, and sends the generated service search request to the server 104 for subsequent service query processing.

[0049] Step 203: read the original service search results corresponding to the query term.

[0050] The search results of service search are documents. The relationship between query and document is similar to the relationship between user and item in product recommendation. The original service search results refer to all the original query results found by the search engine based on the query terms. These original service search results are similar to the input query terms, but may not be relevant. Therefore, it is necessary to filter out the original service search results that are not relevant to obtain the final target service search results.

[0051] Specifically, when performing a service search, the server 104 can search for all original service search results that match the query terms from a preset database based on the keywords corresponding to the query terms. These original service search results are similar to the query terms, but may or may not be relevant. The original service search results related to the query terms are the target service search results required by the user.

[0052] Step 205, identifying the association relationship between the query term and the service knowledge graph of the original service search results.

[0053] Among them, the knowledge graph is a kind of natural language processing technology, which is called knowledge domain visualization or knowledge domain mapping map in the library and information industry. It is a series of various graphics that show the development process and structural relationship of knowledge. It uses visualization technology to describe knowledge resources and their carriers, and mine, analyze, construct, draw and display knowledge and their mutual connections. It is a modern theory that combines the theories and methods of applied mathematics, graphics, information visualization technology, information science and other disciplines with methods such as metrology citation analysis and co-occurrence analysis, and uses visualized graphs to vividly display the core structure, development history, frontier fields and overall knowledge architecture of the discipline to achieve the purpose of multidisciplinary integration. The preset service knowledge graph refers to a pre-constructed, established based on the service attributes of the entity, and is specifically applied to the service search knowledge graph. At present, there is little coverage of service entities in the general knowledge graph. The relationship between general graph knowledge is not suitable for service search. Therefore, this application constructs a knowledge graph framework and a service knowledge graph specifically for services for the scenario of providing services to users. The correlation between the query terms and the original service search results can be mined through the constructed service knowledge graph, that is, the service knowledge graph association relationship between the query terms and the original service search results.

[0054] Specifically, the current search method generally uses semantic models such as CONV-KNRM, BERT, DSSM, CDSSM, Match-pyramid, DRMM, etc. to mine the relationship between the keywords (query) input by the user and the query results (document) during the search process. The semantic model learns the superficial correlation, but cannot learn the deeper knowledge relationship between services. At the same time, when there are multiple levels of hierarchical relationships in the service search data, the calculation correlation of the semantic model is basically invalid. For example, query = junior high school English teaching material tutoring book, document = high school English teaching material tutoring book, because the text is indeed very similar, the semantic model judges that it is related, but it is actually irrelevant. The technical method scheme of the present application can solve this situation where the semantics are completely different due to different "service scopes", which is a good supplement to the traditional semantic matching model. In the search process, by identifying the service knowledge graph association relationship between the query term and the original service search result, it can be determined whether the query term and the original service search result are associated in the multi-level category tree of the knowledge graph, so as to mine their association relationship from a deeper level, thereby improving the search accuracy of the service search process.

[0055] Step 207, extract the data in the original service search results that have a service knowledge graph hierarchy relationship with the query term to obtain the target service search results.

[0056] Among them, the service knowledge graph association relationship specifically refers to the hierarchical relationship in the service knowledge graph. The target service search result is the final service search result corresponding to the query term entered by the user. These service search results are similar to the query terms and are relevant. The service search process can be completed by feeding back the target service search results and the target service search results to terminal 102.

[0057] Specifically, after obtaining the association relationship between the query term and the original service search result, if it is determined based on the preset service knowledge graph that the query term and the original service search result have an association relationship, it means that the original service search result is the target service search result requested by the terminal 102. However, when the query term and the original service search result do not have an association relationship, the original service search result is not the target service search result requested by the terminal 102. By extracting data in the original service search result that has a service knowledge graph upper and lower relationship with the query term, the required target service search result can be effectively obtained from all similar original service search results.

[0058] The above-mentioned service search method obtains the service search request by obtaining a service search request; reading the original service search results corresponding to the query terms; identifying the service knowledge graph association relationship between the query terms and the original service search results; extracting data in the original service search results that have a service knowledge graph hyponymy with the query terms, and obtaining the target service search results. The service search method of the present application obtains the service search request by obtaining the original service search results by identifying the service knowledge graph association relationship between the query terms and the original service search results; and extracting data in the original service search results that have a service knowledge graph hyponymy with the query terms, and obtaining the final target service search results based on these data. The obtained target service search results can better reflect the association relationship with the query terms, and can effectively improve the search accuracy of the service search process.

[0059] In one embodiment, Figure 3 As shown, step 203 includes:

[0060] Step 302: Search for index keywords corresponding to the query terms in a preset index library.

[0061] Step 304: Obtain original service search results corresponding to the query terms based on the index keywords.

[0062] Among them, the preset index library refers to a pre-built index database used to store index directories related to various services. In the process of establishing the search system, vertical search will create a data file similar to a bibliography with the captured information to facilitate high-speed retrieval. The key technical points involved in the index system include word segmentation technology, pre-scoring and post-scoring, incremental indexing and full indexing, sorting technology, hot word cache, standard search statement parsing, etc. Index keywords are keywords in the index directory, which are used to establish the association between the directory and the information. The index keywords can be used to determine the original service search results corresponding to the query terms.

[0063] Specifically, a vertical search engine is a professional search engine for a certain industry. It is a subdivision and extension of the search engine. It integrates a certain type of special information in the web page library, extracts the required data in a targeted manner, processes it, and then returns it to the user in some form. Vertical search is a new search engine service model proposed in contrast to the large amount of information, inaccurate query, and insufficient depth of general search engines. It provides valuable information and related services for a specific field, a specific group of people, or a specific need. Its characteristics are "specialization, precision, and depth", and it has industry characteristics. Compared with the disordered massive information of general search engines, vertical search engines appear to be more focused, specific, and in-depth. Therefore, as a type of vertical search, service search integrates the above characteristics of vertical search. When performing service search, you can first find the directory corresponding to the service scenario corresponding to the query term in the preset index library based on the query term, and then find the index keyword corresponding to the query term from the directory. And obtain the original service search result corresponding to the query term based on the index keyword, so as to effectively obtain the original service search result similar to the query term. In this embodiment, the original service search results corresponding to the query terms are obtained by using the index keywords corresponding to the query terms, which can effectively improve the efficiency of obtaining the original service search results, thereby improving the efficiency of service search.

[0064] In one embodiment, if Figure 4 As shown, step 205 includes:

[0065] Step 401, identifying a first set of entity nodes corresponding to the query term in the preset service knowledge graph, and a second set of entity nodes corresponding to the original service search results in the preset service knowledge graph.

[0066] Step 403, based on the first entity node set, obtain the first hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search result, and based on the second entity node set, obtain the second hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search result.

[0067] Step 405: Identify the service knowledge graph association relationship between the query term and the original service search result based on the first hyponym relationship and the second hyponym relationship.

[0068] Among them, the preset service knowledge graph refers to a knowledge graph pre-constructed for the service field to which the current query term belongs, including various entity nodes and also including the association relationship between various entity nodes. The query term and the original service search result can be placed in the preset service knowledge graph, and then the entity node corresponding to the query term and other entity nodes that form a superior-female relationship with the entity node corresponding to the original service search result are respectively identified, and a first entity node set and a second entity node set are respectively constructed according to the identified entity nodes. The first entity node set and the second entity node set are similar to the set composed of nodes in a multi-level category tree. The association relationship between the query term and the original service search result can specifically refer to the superior-female relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search result, which can specifically include the following types of association relationships: the entity node corresponding to the query term is the superior of the entity node corresponding to the original service search result, the entity node corresponding to the query term is the subordinate of the entity node corresponding to the original service search result, the entity node corresponding to the original service search result is the superior of the entity node corresponding to the query term, and the entity node corresponding to the original service search result is the subordinate of the entity node corresponding to the query term.

[0069] Specifically, the identified association relationship between the query terms and the original service search results in the preset service knowledge graph is specifically the hierarchical relationship between their corresponding entity nodes. In order to identify these hierarchical relationships, the first entity node set corresponding to the query terms in the preset service knowledge graph can be constructed respectively, so as to summarize all hierarchical entities related to the query terms. Similarly, all hierarchical entities related to the current original service search results are summarized through the second entity node set. Then, based on the first entity node set, the first hierarchical relationship between the entity node corresponding to the query terms and the entity node corresponding to the original service search results is obtained. Specifically, it can be determined whether the current original service search results are included in the first entity node set. When the original service search results are included in the first entity node set, it can be determined that there is a corresponding hierarchical relationship between the two. Similarly, the second hierarchical relationship between the entity node corresponding to the query terms and the entity node corresponding to the original service search results can be obtained by determining whether the query terms belong to the second entity node set. Then, by combining the first hierarchical relationship and the second hierarchical relationship, the association relationship between the query terms and the original service search results in the preset service knowledge graph is obtained. Here, the judgment can be made specifically through the logic of "or". As long as any one of the first hierarchical relationship or the second hierarchical relationship determines that the query term and the original service search result have a hierarchical relationship, it can be determined that the query term and the original service search result are associated. In this embodiment, the association between the query term and the original service search result in the preset service knowledge graph is identified through a set of entity nodes, which can effectively improve the accuracy of association relationship identification and ensure the accuracy of service search. As shown in Table 1, in one of the embodiments, it can be obtained through experiments that the hierarchical reasoning based on the knowledge graph can effectively improve the accuracy of service search.

[0070] Table 1: Service search accuracy table

[0071]

[0072] In one embodiment, if Figure 5 As shown, step 401 includes:

[0073] Step 502: perform word segmentation processing on the query terms to obtain word segmentation results.

[0074] Step 504: Obtain the word segmentation combination result corresponding to the word segmentation result of the entry through the N-gram word segmentation algorithm.

[0075] Step 506: query the preset service knowledge graph based on the word segmentation combination result to obtain the first entity node set corresponding to the query term.

[0076] Among them, word segmentation processing specifically refers to the process of recombining a continuous sequence of characters into a word sequence according to certain specifications. The word segmentation processing in this application is specifically word segmentation processing, which can split the query terms input by the user into multiple different words. The word segmentation processing method can specifically include a word segmentation method based on string matching, a word segmentation method based on understanding, and a word segmentation method based on statistics. The N-gram word segmentation algorithm specifically refers to the ngram algorithm. The role of the N-gram word segmentation algorithm is to combine the words in the word sequence obtained by word segmentation in order. The N-gram word segmentation algorithm obtains the 1-gram, 2-gram and 3-gram of the text corresponding to the query term. 1-gram represents the word segmentation combination result composed of a single word, 2-gram represents the word segmentation combination result composed of two word segmentations, and 3-gram represents the word segmentation combination result composed of three word segmentations.

[0077] Specifically, the first entity node set can be constructed by querying the terms and the preset service knowledge graph. The construction process first needs to perform word segmentation on the query terms to obtain the word segmentation results. For example, for the term "junior high school English teaching material tutoring book", the word segmentation algorithm can be used to split it into the word segmentation results of "junior high school", "English", "teaching material", "tutoring" and "book". Then the word segmentation combination result corresponding to the word segmentation result is obtained by the N-ary word segmentation algorithm to obtain as many entities linked to the query terms as possible, and finally the preset service knowledge graph is queried based on the word segmentation combination result to obtain the first entity node set corresponding to the query terms. In this process, it can be specifically based on whether the preset service knowledge graph contains the same words or similar words as the word segmentation combination result, and these same words or similar words are placed in the first entity node set to obtain the first entity node set corresponding to the query terms. In particular, if it is identified that there are overlapping hypernyms and hyponyms in the first entity node set, only the hypernyms of the overlapping part can be retained, and the hyponyms can be removed to distinguish the scope to the greatest extent. Similarly, the corresponding second entity node set can be obtained by performing similar processing to the above processing process on the original service search results. In this embodiment, the query terms are processed through word segmentation and combination processes to effectively construct the entity node set corresponding to the query terms and ensure the validity of the set.

[0078] In one embodiment, if Figure 6 As shown, step 506 includes:

[0079] Step 601, query the preset service knowledge graph based on the word segmentation combination result, and find the entity node whose semantics is consistent with the preset service knowledge graph and the word segmentation combination result.

[0080] Step 603: construct a first entity node set according to the found entity nodes.

[0081] Specifically, the preset service knowledge graph can be regarded as composed of a multi-level category tree, and the preset service knowledge graph contains various entity nodes. At the same time, there are hierarchical relationships between entity nodes. When it is necessary to construct a list of entity nodes corresponding to the query terms, multiple word segmentation combination results can be obtained by splitting and reorganizing the query terms. Then query the preset service knowledge graph to determine whether the preset service knowledge graph includes entity nodes that are semantically consistent with the word segmentation combination results. If included, the word segmentation combination result can be used as an element and added to the first entity node set. In this embodiment, by performing semantic comparison and recognition on the word segmentation combination results, entity nodes can be effectively extracted from the preset service knowledge graph to construct the first entity node set.

[0082] In one of the embodiments, step 403 includes: the first hierarchical relationship includes the existence of a hierarchical relationship and the non-existence of a hierarchical relationship. When the entity nodes in the first entity node set include the entity nodes corresponding to the original service search results, it is determined that the query term has a hierarchical relationship with the original service search results; when the entity nodes in the first entity node set do not include the entity nodes corresponding to the original service search results, it is determined that the query term does not have a hierarchical relationship with the original service search results.

[0083] Specifically, whether the query term and the original service search results have a hierarchical relationship can be determined by whether the entity nodes in the first entity node set include the entity nodes corresponding to the original service search results. When the entity nodes in the first entity node set include the entity nodes corresponding to the original service search results, it is determined that the query term and the original service search results have a hierarchical relationship; when the entity nodes in the first entity node set do not include the entity nodes corresponding to the original service search results, it is determined that the query term and the original service search results do not have a hierarchical relationship. Similarly, similar processing can be performed on the second entity node set. When the nodes in the second entity node set include the entity nodes corresponding to the query term, it is determined that the query term and the original service search results have a hierarchical relationship. In a specific embodiment, the text A (query term) and the text B (original service search results) are respectively linked to entities in the knowledge graph, and their corresponding entity node sets are constructed to obtain the A entity set and the B entity set. In the process of judging their relevance, the relevance Similarity between A and B can be initialized to 0 (irrelevant): then entity A is taken from the A entity set, and the hypernym and hyponym of entity B are taken from the B entity set, and it is judged whether entity A is equal to the hypernym or hyponym of entity B. If there is a hypernym or hyponym of entity B that is equal to entity A, then it is judged that text A (query term) and text B (original service search result) have an association relationship, and Similarity = 1. Similarly, a similar judgment is made for entity B, and finally the final association relationship identification result is determined based on the output Similarity. In this embodiment, the correlation between entity nodes is judged by whether the entity nodes in the first entity node set contain the entity nodes corresponding to the original service search results, which can effectively improve the accuracy of the judgment.

[0084] In one embodiment, if Figure 7 As shown, before step 401, the following steps are also included:

[0085] Step 702, obtaining the service scenario and service entity corresponding to the service search.

[0086] Step 704: Identify service attributes and entity relationships corresponding to the service entities.

[0087] Step 706: construct a preset service knowledge graph corresponding to the service scenario based on the service attributes and entity relationships corresponding to the service entities.

[0088] The service scenario refers to the specific application scenario faced by the service, such as education, medicine, and technology. The service entity refers to all the things that can be involved in the application scenario. The service attribute is specifically related to the actual service content, which can include service entities, service behaviors, service fillings, and service brands. The entity relationship can be a hierarchical relationship or a synonym relationship.

[0089] Specifically, before the method is implemented, a preset service knowledge graph corresponding to each searchable service item may be constructed. For each application scenario, a corresponding preset service knowledge graph needs to be constructed. The preset service knowledge graph is constructed based on service entities and the relationships between service entities. When constructing the preset service knowledge graph, the service entities in the service scenario may be obtained by using technologies such as web crawlers, and then the preset service knowledge graph may be constructed based on the specific attributes of the service entities and the relationships between the service entities. In one embodiment, part of the content of a preset service knowledge graph for a housekeeping service may refer to Figure 8 , the content input by the user, i.e. the housekeeping purpose, includes keywords such as "Brand A home appliance cleaning" and "air conditioner cleaning home service". Through the preset service knowledge graph, it can be identified that "home appliances" and "air conditioners" are service subject words, and there is a hierarchical relationship between them. "Cleaning" and "cleaning" are service behavior words, and they are synonymous. "Home" is a service status word. "Brand A" is a service brand. By constructing a preset service knowledge graph corresponding to the service scenario, the association between the query terms of the service search request and the original service search results can be more effectively identified during the service search process, thereby achieving more accurate service search and improving the accuracy of service search.

[0090] In one embodiment, the service search method of the present application is used to implement educational service search, and the corresponding search process can refer to Fig. 9. In the search process, the text input by the user, that is, the query term is "junior high school English teaching materials guidance book". In the process of constructing the first entity node set corresponding to the query term, it can be determined based on the preset service knowledge graph that "junior high school English" is the hypernym of "junior high school English teaching materials guidance book", and "junior high school" is the hypernym of "junior high school English", and the hypernym of "junior high school" in the preset service knowledge graph is "education". Through retrieval, it can be determined that the original service search results similar to "junior high school English teaching materials guidance book" include "high school English teaching materials guidance book", "high school English teaching materials guidance book" and "high school English teaching materials guidance book". "High school English" is the hypernym of "high school English teaching materials guidance book". And "high school English" is the hypernym of "high school English teaching materials guidance book". "High school English" is the hypernym of "high school English", "high school English" and "high school English teaching materials guidance book". "High school" is the hypernym of "high school English", and the hypernym of "high school" in the preset service knowledge graph is "education". There is no correlation between the terms "junior high school English teaching materials guidance book" and "senior high school English teaching materials guidance book", "senior high school English teaching materials guidance book" and "senior high school English teaching materials guidance book". An ordinary semantic model may determine that these three words are related words, but through the service search method of the present application, it can be determined that these concepts do not have a hierarchical relationship in the service search, that is, "senior high school English teaching materials guidance book", "senior high school English teaching materials guidance book" and "senior high school English teaching materials guidance book" do not belong to the service search results corresponding to "junior high school English teaching materials guidance book".

[0091] It should be understood that although Figure 2-9 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2-9 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0092] In one embodiment, Fig.10 As shown, a vertical search device is provided, which can be a software module or a hardware module, or a combination of the two to form a part of a computer device. The device specifically includes: a request acquisition module 1001, an original result search module 1003, an association relationship identification module 1005 and a target result identification module 1007, wherein:

[0093] The request acquisition module 1001 is used to acquire a service search request, where the service search request carries a query term.

[0094] The original result search module 1003 is used to read the original service search results corresponding to the query term.

[0095] The association relationship identification module 1005 is used to identify the association relationship between the query term and the service knowledge graph of the original service search result.

[0096] The target result identification module 1007 is used to extract data in the original service search results that have a service knowledge graph upper and lower relationship with the query term to obtain the target service search results.

[0097] In one embodiment, the original result search module 1003 is specifically used to: search for index keywords corresponding to the query terms in a preset index library; and obtain original service search results corresponding to the query terms based on the index keywords.

[0098] In one of the embodiments, the association relationship identification module 1005 is specifically used to: identify a first set of entity nodes corresponding to the query term in a preset service knowledge graph, and a second set of entity nodes corresponding to the original service search results in the preset service knowledge graph; based on the first entity node set, obtain a first hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search results, and based on the second entity node set, obtain a second hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search results; based on the first hierarchical relationship and the second hierarchical relationship, identify the service knowledge graph association relationship between the query term and the original service search results.

[0099] In one of the embodiments, the association relationship identification module 1005 is also used to: perform word segmentation processing on the query terms to obtain the word segmentation results; obtain the word segmentation combination results corresponding to the word segmentation results through the N-gram word segmentation algorithm; query the preset service knowledge graph based on the word segmentation combination results to obtain the first entity node set corresponding to the query terms.

[0100] In one of the embodiments, the association relationship identification module 1005 is also used to: query the preset service knowledge graph based on the word segmentation combination result, find the entity nodes whose semantics are consistent between the preset service knowledge graph and the word segmentation combination result; and construct a first entity node set based on the found entity nodes.

[0101] In one of the embodiments, the association relationship identification module 1005 is also used to: when the entity nodes in the first entity node set include the entity nodes corresponding to the original service search results, determine that there is a hierarchical relationship between the query term and the original service search results; when the entity nodes in the first entity node set do not include the entity nodes corresponding to the original service search results, determine that there is no hierarchical relationship between the query term and the original service search results.

[0102] In one of the embodiments, it also includes a knowledge graph construction module, which is used to: obtain service scenarios and service entities corresponding to service searches; identify service attributes and entity relationships corresponding to service entities; and construct a preset service knowledge graph corresponding to the service scenarios based on the service attributes and entity relationships corresponding to the service entities.

[0103] For the specific definition of the vertical search device, please refer to the definition of the service search method above, which will not be repeated here. Each module in the above vertical search device can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0104] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig.11 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store vertical search data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a service search method is implemented.

[0105] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0106] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0107] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0108] In one embodiment, a computer program product or computer program is provided, the computer program product or 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 performs the steps in the above-mentioned method embodiments.

[0109] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0110] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0111] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A service search method, characterized in that: The method comprises: Obtaining a service search request, wherein the service search request carries a query term; Reading the original service search results corresponding to the query term; Identify a first entity node set corresponding to the query term in a preset service knowledge graph, and a second entity node set corresponding to the original service search result in the preset service knowledge graph; the first entity node set is constructed by other entity nodes that form a superior-female relationship with the entity node corresponding to the query term, and the second entity node set is constructed by other entity nodes that form a superior-female relationship with the entity node corresponding to the original service search result; Based on the first entity node set, a first hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search result is obtained; based on the second entity node set, a second hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search result is obtained; the first hierarchical relationship is determined by whether the original service search result is included in the first entity node set; the second hierarchical relationship is determined by whether the query term is included in the second entity node set; Determining that the query term and the original service search result have a hierarchical relationship in any one of the first hierarchical relationship and the second hierarchical relationship, and determining that the query term and the original service search result are associated; The data associated with the query term in the original service search results is extracted to obtain the target service search results.

2. The method according to claim 1, characterized in that The reading of the original service search result corresponding to the query term includes: Searching for index keywords corresponding to the query terms in a preset index library; The original service search result corresponding to the query term is read based on the index keyword.

3. The method according to claim 1, characterized in that The step of identifying the first entity node set corresponding to the query term in the preset service knowledge graph includes: Performing word segmentation processing on the query terms to obtain word segmentation results; Obtain the word segmentation combination result corresponding to the word segmentation result of the term through the N-gram word segmentation algorithm; The preset service knowledge graph is queried based on the word segmentation combination result to obtain a first entity node set corresponding to the query term.

4. The method according to claim 3, characterized in that The querying the preset service knowledge graph based on the word segmentation combination result to obtain the first entity node set corresponding to the query term includes: Based on the word segmentation combination result, the preset service knowledge graph is queried to find an entity node in the preset service knowledge graph that is semantically consistent with the word segmentation combination result; A first entity node set is constructed according to the found entity nodes.

5. The method according to claim 1, characterized in that The first hierarchical relationship includes the existence of a hierarchical relationship and the non-existence of a hierarchical relationship. The acquiring, based on the first entity node set, a first hierarchical relationship between an entity node corresponding to the query term and an entity node corresponding to the original service search result comprises: When the entity nodes in the first entity node set include the entity node corresponding to the original service search result, determining that the query term and the original service search result have a hierarchical relationship; When the entity nodes in the first entity node set do not include the entity node corresponding to the original service search result, it is determined that there is no hierarchical relationship between the query term and the original service search result.

6. The method according to claim 1, characterized in that Before identifying the service knowledge graph association relationship between the query term and the original service search result, the method further includes: Get the service scenario and service entity corresponding to the service search; Identify service attributes and entity relationships corresponding to the service entity; Based on the service attributes and entity relationships corresponding to the service entities, a preset service knowledge graph corresponding to the service scenario is constructed.

7. A service search device, characterized in that: The device comprises: A request acquisition module, used to acquire a service search request, wherein the service search request carries a query term; An original result search module, used to read the original service search results corresponding to the query term; An association relationship identification module is used to identify a first entity node set corresponding to the query term in a preset service knowledge graph, and a second entity node set corresponding to the original service search result in the preset service knowledge graph; the first entity node set is constructed by other entity nodes that form a hierarchical relationship with the entity node corresponding to the query term, and the second entity node set is constructed by other entity nodes that form a hierarchical relationship with the entity node corresponding to the original service search result; based on the first entity node set, a first hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search result is obtained, and based on the second entity node set, a second hierarchical relationship between the entity node corresponding to the query term and the entity node corresponding to the original service search result is obtained; the first hierarchical relationship is determined by whether the original service search result is included in the first entity node set; the second hierarchical relationship is determined by whether the query term is included in the second entity node set; in any one of the first hierarchical relationship and the second hierarchical relationship, it is determined that the query term and the original service search result have a hierarchical relationship, and the query term and the original service search result are associated; The target result identification module is used to extract data associated with the query term in the original service search results to obtain the target service search results.

8. The device according to claim 7, characterized in that The original result search module is specifically used to: search for an index keyword corresponding to the query term in a preset index library; and read the original service search result corresponding to the query term based on the index keyword.

9. The device according to claim 7, characterized in that The association relationship identification module is also used to perform word segmentation processing on the query terms to obtain word segmentation results; obtain word segmentation combination results corresponding to the word segmentation results through an N-gram word segmentation algorithm; and query the preset service knowledge graph based on the word segmentation combination results to obtain a first entity node set corresponding to the query terms.

10. The device according to claim 9, characterized in that The association relationship identification module is also used to: query the preset service knowledge graph based on the word segmentation combination result, find the entity nodes in the preset service knowledge graph that are semantically consistent with the word segmentation combination result; and construct a first entity node set based on the found entity nodes.

11. The device according to claim 7, characterized in that The first hierarchical relationship includes the existence of a hierarchical relationship and the non-existence of a hierarchical relationship. The association relationship identification module is also used to: when the entity nodes in the first entity node set include the entity nodes corresponding to the original service search results, determine that the query term has a hierarchical relationship with the original service search results; when the entity nodes in the first entity node set do not include the entity nodes corresponding to the original service search results, determine that the query term does not have a hierarchical relationship with the original service search results.

12. The device according to claim 7, characterized in that The device also includes a knowledge graph construction module; The knowledge graph construction module is used to: obtain the service scenario and service entity corresponding to the service search; identify the service attributes and entity relationships corresponding to the service entity; Based on the service attributes and entity relationships corresponding to the service entities, a preset service knowledge graph corresponding to the service scenario is constructed.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

14. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images