An efficient retrieval method and system for structured data based on knowledge graph

Abstract

The present invention is applicable to the field of knowledge graph technology, and provides a method and system for efficient retrieval of structured data based on knowledge graph. The present invention receives a comprehensive retrieval request from a user, analyzes the comprehensive retrieval request, and determines whether an efficient retrieval is required; when an efficient retrieval is required, obtains multiple limited retrieval evaluation data; performs relevant and irrelevant comprehensive evaluation analysis, calculates multiple recommended evaluation values; compares multiple recommended evaluation values, selects a recommended retrieval platform; performs efficient retrieval processing on multiple retrieval key information, and generates and displays retrieval result information. When an efficient retrieval is required, it is possible to obtain multiple limited retrieval evaluation data, calculate and compare multiple recommended evaluation values, select a recommended retrieval platform from multiple knowledge graph data retrieval platforms for efficient retrieval processing, and automatically select the most suitable recommended retrieval platform for the user, effectively improving the accuracy of data retrieval.

Description

Technical Field

[0001] The present invention belongs to the technical field of knowledge graphs, and in particular, relates to an efficient retrieval method and system for structured data based on knowledge graphs. Background Art

[0002] A knowledge graph is a structured semantic knowledge base that is used to describe concepts in the physical world and their relationships in symbolic form.

[0003] Efficient retrieval of knowledge graphs uses the structured characteristics of knowledge graphs and specific retrieval algorithms and technologies to achieve fast, accurate, and comprehensive retrieval of data. Efficient retrieval of knowledge graphs has a wide range of application scenarios in many fields, such as search engines, intelligent question-answering systems, and recommendation systems.

[0004] At present, with the rapid development of knowledge graph retrieval technology, numerous knowledge graph retrieval platforms have emerged and are widely used. Since different knowledge graph retrieval platforms have different databases, when users pursue efficient and accurate retrieval experience, it is difficult to choose the most suitable platform for efficient retrieval from the numerous knowledge graph retrieval platforms, which affects the accuracy of data retrieval. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a method and system for efficient retrieval of structured data based on a knowledge graph, aiming to solve the technical problems existing in the prior art mentioned in the background technology.

[0006] The embodiment of the present invention is implemented as follows:

[0007] A method for efficient retrieval of structured data based on knowledge graph, the method specifically comprising the following steps:

[0008] Receive a comprehensive search request from a user, analyze the comprehensive search request, obtain target search requirements and multiple key search information, and determine whether an efficient search is required;

[0009] When efficient retrieval is required, limited retrieval evaluation data is obtained from multiple preset knowledge graph data retrieval platforms;

[0010] Performing relevant and irrelevant comprehensive evaluation analysis on the plurality of limited search evaluation data, and calculating the recommended evaluation values ​​corresponding to the plurality of knowledge graph data retrieval platforms;

[0011] Comparing the plurality of recommendation evaluation values, and selecting a recommended retrieval platform from the plurality of knowledge graph data retrieval platforms;

[0012] Based on the recommendation search platform, efficient search processing is performed on multiple search key information to generate and display search result information.

[0013] As a further limitation of the technical solution of the embodiment of the present invention, the receiving of the comprehensive search request of the user, analyzing the comprehensive search request, obtaining the target search requirement and multiple key search information, and determining whether an efficient search is required specifically includes the following steps:

[0014] Perform request interaction in the preset search request interface and receive the user's comprehensive search request;

[0015] Analyze the comprehensive search request to obtain search request information and target search requirements;

[0016] Performing key identification on the search request information, and extracting a plurality of key search information from the search request information;

[0017] The target search requirement is identified to determine whether an efficient search is required.

[0018] As a further limitation of the technical solution of the embodiment of the present invention, when efficient retrieval is required, obtaining limited retrieval evaluation data from multiple preset knowledge graph data retrieval platforms specifically includes the following steps:

[0019] When efficient retrieval is required, historical retrieval record data is obtained from multiple preset knowledge graph data retrieval platforms;

[0020] Performing evaluation, identification and screening on the plurality of historical search record data to generate a plurality of historical search evaluation data;

[0021] According to the preset limited quantity requirement, the latest limited search evaluation data are screened from the plurality of historical search evaluation data.

[0022] As a further limitation of the technical solution of the embodiment of the present invention, the performing of relevant and irrelevant comprehensive evaluation analysis on the plurality of the limited search evaluation data and calculating the recommended evaluation values ​​corresponding to the plurality of the knowledge graph data retrieval platforms specifically includes the following steps:

[0023] Relevant historical evaluation identification is performed on the plurality of limited search evaluation data respectively, and a plurality of relevant evaluation results are obtained;

[0024] Identifying irrelevant historical evaluations for the plurality of limited search evaluation data respectively, and obtaining a plurality of irrelevant evaluation results;

[0025] Import preset coefficient setting data;

[0026] According to the coefficient setting data, the multiple relevant evaluation results and the multiple irrelevant evaluation results, the recommended evaluation values ​​corresponding to the multiple knowledge graph data retrieval platforms are calculated.

[0027] As a further limitation of the technical solution of the embodiment of the present invention, the calculation formula of the recommendation evaluation value corresponding to the multiple knowledge graph data retrieval platforms is:

[0028] ;

[0029] ;

[0030] in, Representative A knowledge graph data retrieval platform, For the The recommended evaluation value of the knowledge graph data retrieval platform, is the correlation coefficient, is the irrelevant coefficient, Representative Related evaluation results, For the The total number of relevant evaluation results of knowledge graph data retrieval platforms, is the recommended coefficient, For the The evaluation values ​​of the relevant evaluation results are assigned. Representative Irrelevant evaluation results, For the The total number of irrelevant evaluation results of knowledge graph data retrieval platforms, For the The evaluation assignment is irrelevant to the evaluation results. is the standardized coefficient.

[0031] As a further limitation of the technical solution of the embodiment of the present invention, the comparing of the plurality of recommendation evaluation values ​​and selecting the recommendation retrieval platform from the plurality of knowledge graph data retrieval platforms specifically comprises the following steps:

[0032] Comparing the plurality of recommendation evaluation values ​​to generate a recommendation comparison result;

[0033] According to the recommendation comparison result, selecting the best evaluation value from the plurality of recommended evaluation values;

[0034] From the multiple knowledge graph data retrieval platforms, select a recommended retrieval platform corresponding to the best evaluation value.

[0035] As a further limitation of the technical solution of the embodiment of the present invention, the efficient retrieval processing of the multiple retrieval key information based on the recommendation retrieval platform, generating and displaying the retrieval result information specifically includes the following steps:

[0036] Based on the recommendation search platform, a plurality of the key search information are efficiently searched and processed to generate search result information;

[0037] Create a search display interface;

[0038] In the search display interface, the search result information is displayed.

[0039] A knowledge graph-based efficient retrieval system for structured data, the system comprising a retrieval request analysis module, a limited data acquisition module, a comprehensive evaluation analysis module, a recommendation platform selection module and an efficient retrieval display module, wherein:

[0040] A search request analysis module is used to receive a comprehensive search request from a user, analyze the comprehensive search request, obtain a target search requirement and multiple key search information, and determine whether an efficient search is required;

[0041] A limited data acquisition module is used to obtain limited search evaluation data from multiple preset knowledge graph data retrieval platforms when efficient retrieval is required;

[0042] A comprehensive evaluation analysis module, used to perform relevant and irrelevant comprehensive evaluation analysis on the plurality of limited search evaluation data, and calculate the recommended evaluation values ​​corresponding to the plurality of knowledge graph data retrieval platforms;

[0043] A recommendation platform selection module, used to compare the plurality of recommendation evaluation values ​​and select a recommendation retrieval platform from the plurality of knowledge graph data retrieval platforms;

[0044] The efficient search and display module is used to perform efficient search processing on multiple search key information based on the recommended search platform, and generate and display search result information.

[0045] As a further limitation of the technical solution of the embodiment of the present invention, the search request analysis module specifically includes:

[0046] A request interaction unit, used to perform request interaction in a preset search request interface and receive a user's comprehensive search request;

[0047] A request analysis unit, used to analyze the comprehensive search request to obtain search request information and target search requirements;

[0048] A key identification unit, used to perform key identification on the search request information, and extract multiple search key information from the search request information;

[0049] The requirement judgment unit is used to identify the target search requirement and judge whether an efficient search is required.

[0050] As a further limitation of the technical solution of the embodiment of the present invention, the limited data acquisition module specifically includes:

[0051] A record acquisition unit, used to acquire historical search record data from multiple preset knowledge graph data retrieval platforms when efficient retrieval is required;

[0052] An evaluation and screening unit, used for performing evaluation, identification and screening on the plurality of historical search record data to generate a plurality of historical search evaluation data;

[0053] The limited screening unit is used to screen the latest limited retrieval evaluation data from the multiple historical retrieval evaluation data according to the preset limited quantity requirements.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The embodiment of the present invention receives a comprehensive search request from a user, analyzes the comprehensive search request, and determines whether an efficient search is required; when an efficient search is required, obtains multiple limited search evaluation data; performs relevant and irrelevant comprehensive evaluation analysis, calculates multiple recommended evaluation values; compares multiple recommended evaluation values, selects a recommended search platform; performs efficient search processing on multiple search key information, and generates and displays search result information. When an efficient search is required, it can obtain multiple limited search evaluation data, calculate and compare multiple recommended evaluation values, select a recommended search platform from multiple knowledge graph data retrieval platforms for efficient search processing, and automatically select the most suitable recommended search platform for the user, effectively improving the accuracy of data retrieval. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flowchart of an efficient structured data retrieval method based on a knowledge graph provided by an embodiment of the present invention is shown;

[0057] Figure 2 A flowchart showing the comprehensive search request reception and analysis in the method provided by an embodiment of the present invention is shown;

[0058] Figure 3 A flowchart of obtaining limited search evaluation data in the method provided by an embodiment of the present invention is shown;

[0059] Figure 4 A flowchart of comprehensive evaluation analysis and calculation in the method provided by an embodiment of the present invention is shown;

[0060] Figure 5 A flowchart of selecting a recommended search platform in the method provided by an embodiment of the present invention is shown;

[0061] Figure 6A flowchart showing efficient retrieval processing in the method provided by an embodiment of the present invention is shown;

[0062] Figure 7 An application architecture diagram of a knowledge graph-based efficient structured data retrieval system provided by an embodiment of the present invention is shown;

[0063] Figure 8 It shows a structural block diagram of a search request analysis module in a system provided by an embodiment of the present invention;

[0064] Fig. 9 The structure block diagram of the limited data acquisition module in the system provided by the embodiment of the present invention is shown. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention 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 invention and are not intended to limit the present invention.

[0066] It is understandable that, at present, with the rapid development of knowledge graph retrieval technology, numerous knowledge graph retrieval platforms have emerged and are widely used. Since different knowledge graph retrieval platforms have different databases, when users pursue efficient and accurate retrieval experience, it is difficult to choose the most suitable platform from the numerous knowledge graph retrieval platforms for efficient retrieval, which affects the accuracy of data retrieval.

[0067] To solve the above problems, the embodiment of the present invention discloses a method and system for efficient retrieval of structured data based on knowledge graph, which receives a comprehensive retrieval request from a user, analyzes the comprehensive retrieval request, obtains target retrieval requirements and multiple retrieval key information, and determines whether efficient retrieval is needed; when efficient retrieval is needed, obtains limited retrieval evaluation data from multiple preset knowledge graph data retrieval platforms; performs relevant and irrelevant comprehensive evaluation analysis on multiple limited retrieval evaluation data, calculates recommended evaluation values ​​corresponding to multiple knowledge graph data retrieval platforms; compares multiple recommended evaluation values, selects a recommended retrieval platform from multiple knowledge graph data retrieval platforms; based on the recommended retrieval platform, performs efficient retrieval processing on multiple retrieval key information, generates and displays retrieval result information. When efficient retrieval is needed, multiple limited retrieval evaluation data can be obtained, multiple recommended evaluation values ​​can be calculated and compared, and a recommended retrieval platform can be selected from multiple knowledge graph data retrieval platforms for efficient retrieval processing, and the most suitable recommended retrieval platform can be automatically selected for the user, effectively improving the accuracy of data retrieval.

[0068] Specifically, Figure 1 A flowchart of an efficient method for retrieval of structured data based on a knowledge graph provided in an embodiment of the present invention is shown.

[0069] In a preferred embodiment of the present invention, a method for efficient retrieval of structured data based on a knowledge graph is provided, and the method specifically comprises the following steps:

[0070] Step S101: receiving a comprehensive search request from a user, analyzing the comprehensive search request, obtaining a target search requirement and multiple key search information, and determining whether an efficient search is required.

[0071] In an embodiment of the present invention, when a user has a need for data retrieval, in a preset retrieval request interface, an information input request interaction is performed with the user, the user is guided to input information and set requirements on the retrieval request interface, and then the user's comprehensive retrieval request is obtained, and the comprehensive retrieval request is analyzed to obtain retrieval request information and target retrieval requirements, and then key identification is performed on the retrieval request information to eliminate retrieval-irrelevant information including modal particles, repeated content, first person, etc., and multiple retrieval key information is extracted from the retrieval request information, and the target retrieval requirements are identified to determine whether an efficient retrieval is required.

[0072] It is understandable that the request for interaction may be carried out in the form of a pop-up window display or voice playback.

[0073] It can be understood that the search request information is the information entered by the user on the search request interface, which is composed of multiple search-irrelevant information and multiple search key information; the target search requirement is generated by the user setting the search requirement on the search request interface. In an embodiment of the present invention, the user can check the box to determine whether an efficient search is required.

[0074] Specifically, Figure 2 A flow chart of comprehensive search request reception and analysis in the method provided by an embodiment of the present invention is shown.

[0075] Among them, in the preferred embodiment provided by the present invention, the receiving of the comprehensive search request of the user, analyzing the comprehensive search request, obtaining the target search requirement and multiple key search information, and determining whether an efficient search is required specifically includes the following steps:

[0076] Step S1011: perform request interaction in a preset search request interface to receive a comprehensive search request from the user.

[0077] Step S1012: Analyze the comprehensive search request to obtain search request information and target search requirements.

[0078] Step S1013: perform key identification on the search request information, and extract multiple search key information from the search request information.

[0079] Step S1014: Identify the target search requirement and determine whether an efficient search is required.

[0080] Furthermore, the efficient retrieval method for structured data based on knowledge graph also includes the following steps:

[0081] Step S102: When efficient retrieval is required, limited retrieval evaluation data is obtained from multiple preset knowledge graph data retrieval platforms.

[0082] In an embodiment of the present invention, when efficient retrieval is required, historical retrieval record data of multiple preset knowledge graph data retrieval platforms are obtained, and multiple historical retrieval record data are identified, records without retrieval personnel evaluations are eliminated, and records with retrieval personnel evaluations are retained, thereby generating multiple historical retrieval evaluation data, and then the multiple historical retrieval evaluation data are filtered by time and quantity according to the time sequence and the preset limited quantity requirements, and the latest limited retrieval evaluation data that meets the limited quantity requirements are filtered out from the multiple historical retrieval evaluation data.

[0083] It is understandable that when it is determined that efficient retrieval is not necessary, a knowledge graph data retrieval platform is randomly selected to perform ordinary retrieval processing on multiple key retrieval information.

[0084] It is understandable that multiple preset knowledge graph data retrieval platforms may include Antuo Knowledge Graph Platform, Gemini Platform, Tongdun Cloud Map, Baifendian Technology Platform, NRD Studio, etc.

[0085] Specifically, Figure 3 A flow chart of obtaining limited search evaluation data in the method provided by an embodiment of the present invention is shown.

[0086] Among them, in the preferred embodiment provided by the present invention, when efficient retrieval is required, obtaining limited retrieval evaluation data from multiple preset knowledge graph data retrieval platforms specifically includes the following steps:

[0087] Step S1021: When efficient retrieval is required, historical retrieval record data is obtained from multiple preset knowledge graph data retrieval platforms.

[0088] Step S1022: Evaluate, identify and screen the plurality of historical search record data to generate a plurality of historical search evaluation data.

[0089] Step S1023: According to the preset limited quantity requirement, the latest limited search evaluation data are selected from the plurality of historical search evaluation data.

[0090] Furthermore, the efficient retrieval method for structured data based on knowledge graph also includes the following steps:

[0091] Step S103: Perform relevant and irrelevant comprehensive evaluation analysis on the multiple limited search evaluation data, and calculate the recommended evaluation values ​​corresponding to the multiple knowledge graph data retrieval platforms.

[0092] In an embodiment of the present invention, multiple search key information is identified to determine the search related type, and then, according to the search related type, the multiple limited search evaluation data are respectively identified with related historical evaluations, and multiple related evaluation results with the same search related type in each limited search evaluation data are determined. At the same time, multiple irrelevant evaluation results with different search related types in each limited search evaluation data are determined, and preset coefficient setting data is imported. Then, according to the coefficient setting data, multiple related evaluation results and multiple irrelevant evaluation results, the recommended evaluation values ​​corresponding to the multiple knowledge graph data retrieval platforms are calculated. Specifically, the calculation formula of the multiple recommended evaluation values ​​is:

[0093] ;

[0094] ;

[0095] in, Representative A knowledge graph data retrieval platform, For the The recommended evaluation value of the knowledge graph data retrieval platform, is the correlation coefficient, is the irrelevant coefficient, Representative Related evaluation results, For the The total number of relevant evaluation results of knowledge graph data retrieval platforms, is the recommended coefficient, For the The evaluation value of the relevant evaluation results is assigned. If the relevant evaluation result is a good one, then ; If the relevant evaluation result is a negative review, then , Representative Irrelevant evaluation results, For the The total number of irrelevant evaluation results of knowledge graph data retrieval platforms, For the The evaluation value of an irrelevant evaluation result is assigned. If the irrelevant evaluation result is a positive evaluation, then , if the irrelevant evaluation result is a bad review, then , is the standardized coefficient.

[0096] It can be understood that the search-related types are classification types determined through comprehensive identification of multiple key search information, and may include different types such as education, medical care, industry, tourism, law, and catering.

[0097] It can be understood that each limited search evaluation data contains multiple search evaluation records, and each search evaluation record can be classified and identified, and the evaluation results corresponding to the search evaluation records belonging to the search-related type can be determined as relevant evaluation results, while the evaluation results corresponding to the search evaluation records that do not belong to the search-related type can be determined as irrelevant evaluation results.

[0098] It can be understood that the coefficient setting data contains specific values ​​of the correlation coefficient, irrelevant coefficient, recommended coefficient and standardized coefficient.

[0099] Specifically, Figure 4 A flow chart of comprehensive evaluation analysis and calculation in the method provided in an embodiment of the present invention is shown.

[0100] Among them, in the preferred embodiment provided by the present invention, the comprehensive evaluation analysis of the relevant and irrelevant data on the multiple limited search evaluation data and the calculation of the recommended evaluation values ​​corresponding to the multiple knowledge graph data retrieval platforms specifically include the following steps:

[0101] Step S1031 : performing relevant historical evaluation identification on the plurality of limited search evaluation data respectively, and obtaining a plurality of relevant evaluation results.

[0102] Step S1032: Identify irrelevant historical evaluations for each of the plurality of limited search evaluation data, and obtain a plurality of irrelevant evaluation results.

[0103] Step S1033: import preset coefficient setting data.

[0104] Step S1034: Calculate the recommended evaluation values ​​corresponding to the multiple knowledge graph data retrieval platforms according to the coefficient setting data, the multiple relevant evaluation results and the multiple irrelevant evaluation results.

[0105] Furthermore, the efficient retrieval method for structured data based on knowledge graph also includes the following steps:

[0106] Step S104: compare the multiple recommendation evaluation values ​​and select a recommended retrieval platform from the multiple knowledge graph data retrieval platforms.

[0107] In an embodiment of the present invention, a recommendation comparison result is generated by comparing the numerical values ​​of multiple recommendation evaluation values, and then according to the recommendation comparison result, the best evaluation value with the largest numerical value is selected from the multiple recommendation evaluation values, and then the recommendation retrieval platform corresponding to the best evaluation value is selected from multiple knowledge graph data retrieval platforms.

[0108] Specifically, Figure 5 A flow chart of selecting a recommendation retrieval platform in the method provided by an embodiment of the present invention is shown.

[0109] Among them, in the preferred embodiment provided by the present invention, the comparing of the plurality of the recommendation evaluation values ​​and selecting the recommended retrieval platform from the plurality of the knowledge graph data retrieval platforms specifically comprises the following steps:

[0110] Step S1041: compare the multiple recommendation evaluation values ​​to generate a recommendation comparison result.

[0111] Step S1042: Select the best evaluation value from the multiple recommended evaluation values ​​according to the recommendation comparison result.

[0112] Step S1043: Select a recommended search platform corresponding to the best evaluation value from the multiple knowledge graph data retrieval platforms.

[0113] Furthermore, the efficient retrieval method for structured data based on knowledge graph also includes the following steps:

[0114] Step S105: Based on the recommendation search platform, perform efficient search processing on the multiple search key information, generate and display search result information.

[0115] In the embodiment of the present invention, based on the recommendation search platform, multiple key search information are efficiently searched and processed to generate search result information, and a search display interface is created, and then the search result information is displayed in the search display interface.

[0116] Specifically, Figure 6 A flow chart showing efficient retrieval processing in the method provided by an embodiment of the present invention is shown.

[0117] Among them, in the preferred embodiment provided by the present invention, the efficient search processing of the multiple search key information based on the recommendation search platform, generating and displaying the search result information specifically includes the following steps:

[0118] Step S1051: Based on the recommendation search platform, perform efficient search processing on the multiple search key information to generate search result information.

[0119] Step S1052: Create a search display interface.

[0120] Step S1053: Display the search result information in the search display interface.

[0121] Furthermore, Figure 7 The application architecture diagram of the knowledge graph-based efficient retrieval system for structured data provided by an embodiment of the present invention is shown.

[0122] Among them, in another preferred embodiment provided by the present invention, an efficient retrieval system for structured data based on knowledge graph includes:

[0123] The search request analysis module 101 is used to receive a comprehensive search request from a user, analyze the comprehensive search request, obtain a target search requirement and multiple key search information, and determine whether an efficient search is required.

[0124] In an embodiment of the present invention, when a user has a need for data retrieval, the retrieval request analysis module 101 interacts with the user's information input request in a preset retrieval request interface, guides the user to input information and set requirements on the retrieval request interface, and then obtains the user's comprehensive retrieval request, obtains retrieval request information and target retrieval requirements by analyzing the comprehensive retrieval request, and then performs key identification on the retrieval request information, eliminates retrieval-irrelevant information including modal particles, repeated content, first person, etc., extracts multiple retrieval key information from the retrieval request information, and identifies the target retrieval requirements to determine whether an efficient retrieval is required.

[0125] Specifically, Figure 8 It shows a structural block diagram of the search request analysis module 101 in the system provided by the embodiment of the present invention.

[0126] Among them, in the preferred implementation mode provided by the present invention, the search request analysis module 101 specifically includes:

[0127] The request interaction unit 1011 is used to perform request interaction in a preset search request interface and receive a comprehensive search request from a user.

[0128] The request analysis unit 1012 is used to analyze the comprehensive search request to obtain the search request information and the target search requirement.

[0129] The key identification unit 1013 is used to perform key identification on the search request information and extract multiple search key information from the search request information.

[0130] The requirement determination unit 1014 is used to identify the target search requirement and determine whether an efficient search is required.

[0131] Furthermore, the efficient retrieval system for structured data based on knowledge graph also includes:

[0132] The limited data acquisition module 102 is used to obtain limited search evaluation data from multiple preset knowledge graph data retrieval platforms when efficient retrieval is required.

[0133] In an embodiment of the present invention, when efficient retrieval is required, the limited data acquisition module 102 obtains historical retrieval record data of multiple preset knowledge graph data retrieval platforms, and identifies multiple historical retrieval record data, eliminates records without retrieval personnel evaluation, and retains records with retrieval personnel evaluation, thereby generating multiple historical retrieval evaluation data, and then performs time screening and quantity screening on the multiple historical retrieval evaluation data in chronological order and according to preset limited quantity requirements, and screens the latest limited retrieval evaluation data that meets the limited quantity requirements from the multiple historical retrieval evaluation data.

[0134] Specifically, Fig. 9 It shows a structural block diagram of the limited data acquisition module 102 in the system provided by the embodiment of the present invention.

[0135] Among them, in the preferred embodiment provided by the present invention, the limited data acquisition module 102 specifically includes:

[0136] The record acquisition unit 1021 is used to obtain historical search record data from multiple preset knowledge graph data retrieval platforms when efficient retrieval is required.

[0137] The evaluation and screening unit 1022 is used to perform evaluation, identification and screening on the plurality of historical search record data to generate a plurality of historical search evaluation data.

[0138] The limited screening unit 1023 is used to screen the latest limited retrieval evaluation data from the plurality of historical retrieval evaluation data according to a preset limited quantity requirement.

[0139] Furthermore, the efficient retrieval system for structured data based on knowledge graph also includes:

[0140] The comprehensive evaluation analysis module 103 is used to perform relevant and irrelevant comprehensive evaluation analysis on the multiple limited search evaluation data, and calculate the recommended evaluation values ​​corresponding to the multiple knowledge graph data retrieval platforms.

[0141] In an embodiment of the present invention, the comprehensive evaluation analysis module 103 identifies multiple search key information, determines the search related type, and then performs relevant historical evaluation identification on multiple limited search evaluation data according to the search related type, determines multiple relevant evaluation results with the same search related type in each limited search evaluation data, and at the same time, determines multiple irrelevant evaluation results with different search related types in each limited search evaluation data, and imports preset coefficient setting data, and then calculates the recommended evaluation values ​​corresponding to multiple knowledge graph data retrieval platforms according to the coefficient setting data, multiple relevant evaluation results and multiple irrelevant evaluation results. Specifically, the calculation formula of multiple recommended evaluation values ​​is:

[0142] ;

[0143] ;

[0144] in, Representative A knowledge graph data retrieval platform, For the The recommended evaluation value of the knowledge graph data retrieval platform, is the correlation coefficient, is the irrelevant coefficient, Representative Related evaluation results, For the The total number of relevant evaluation results of knowledge graph data retrieval platforms, is the recommended coefficient, For the The evaluation value of the relevant evaluation results is assigned. If the relevant evaluation result is a good one, then ; If the relevant evaluation result is a negative review, then , Representative Irrelevant evaluation results, For the The total number of irrelevant evaluation results of knowledge graph data retrieval platforms, For the The evaluation value of an irrelevant evaluation result is assigned. If the irrelevant evaluation result is a positive evaluation, then , if the irrelevant evaluation result is a bad review, then , is the standardized coefficient.

[0145] The recommendation platform selection module 104 is used to compare the multiple recommendation evaluation values ​​and select a recommendation retrieval platform from the multiple knowledge graph data retrieval platforms.

[0146] In an embodiment of the present invention, the recommendation platform selection module 104 generates a recommendation comparison result by comparing the values ​​of multiple recommendation evaluation values, and then selects the best evaluation value with the largest value from the multiple recommendation evaluation values ​​according to the recommendation comparison result, and then selects the recommendation retrieval platform corresponding to the best evaluation value from multiple knowledge graph data retrieval platforms.

[0147] The efficient search and display module 105 is used to perform efficient search processing on the multiple search key information based on the recommendation search platform, and generate and display the search result information.

[0148] In the embodiment of the present invention, the efficient search and display module 105 performs efficient search processing on multiple key search information based on the recommended search platform, generates search result information, creates a search display interface, and then displays the search result information in the search display interface.

[0149] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0150] Those skilled 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 program can be stored in a non-volatile computer-readable storage medium. When the 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 may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0151] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An efficient retrieval method for structured data based on knowledge graph, characterized in that: The method specifically comprises the following steps: Receive a comprehensive search request from a user, analyze the comprehensive search request, obtain target search requirements and multiple key search information, and determine whether an efficient search is required; When efficient retrieval is required, limited retrieval evaluation data is obtained from multiple preset knowledge graph data retrieval platforms; Performing relevant and irrelevant comprehensive evaluation analysis on the plurality of limited search evaluation data, and calculating the recommended evaluation values ​​corresponding to the plurality of knowledge graph data retrieval platforms; Identify multiple search key information, determine the search related types, and then identify the relevant historical evaluations of multiple limited search evaluation data according to the search related types, determine multiple relevant evaluation results with the same search related types in each limited search evaluation data, and at the same time, determine multiple irrelevant evaluation results with different search related types in each limited search evaluation data, and import preset coefficient setting data, and then calculate the recommended evaluation values ​​corresponding to multiple knowledge graph data retrieval platforms according to the coefficient setting data, multiple relevant evaluation results and multiple irrelevant evaluation results; Comparing the plurality of recommendation evaluation values, and selecting a recommended retrieval platform from the plurality of knowledge graph data retrieval platforms; Based on the recommendation search platform, perform efficient search processing on the multiple search key information, generate and display search result information; When efficient retrieval is required, obtaining limited retrieval evaluation data from multiple preset knowledge graph data retrieval platforms specifically includes the following steps: When efficient retrieval is required, historical retrieval record data is obtained from multiple preset knowledge graph data retrieval platforms; Performing evaluation, identification and screening on the plurality of historical search record data to generate a plurality of historical search evaluation data; According to the preset limited quantity requirement, the latest limited search evaluation data are screened from the plurality of historical search evaluation data.

2. The efficient retrieval method for structured data based on knowledge graph according to claim 1 is characterized in that: The receiving of the comprehensive search request from the user, analyzing the comprehensive search request, obtaining the target search requirement and multiple key search information, and determining whether an efficient search is required specifically includes the following steps: Perform request interaction in the preset search request interface and receive the user's comprehensive search request; Analyze the comprehensive search request to obtain search request information and target search requirements; Performing key identification on the search request information, and extracting a plurality of key search information from the search request information; The target search requirement is identified to determine whether an efficient search is required.

3. The efficient retrieval method for structured data based on knowledge graph according to claim 1 is characterized in that: The calculation formula for the recommended evaluation values ​​corresponding to the multiple knowledge graph data retrieval platforms is: {R}_{x}=\left [ {k\times \sum{{}^{{n}_{x}}_{i=1}}\frac {\left ( {{a}^{r\left ( {i} \right )}-1} \right )} {log\left ( {1+i} \right )}+d\times \sum{{}^{{m}_{x}}_{j=1}\frac {\left ( {{a}^{r\left ( {j} \right )}-1} \right )} {log\left ( {1+j} \right )}}} \right ]\times S ; ； in, Representative A knowledge graph data retrieval platform, For the The recommended evaluation value of the knowledge graph data retrieval platform, is the correlation coefficient, is the irrelevant coefficient, Representative Related evaluation results, For the The total number of relevant evaluation results of knowledge graph data retrieval platforms, is the recommended coefficient, For the The evaluation values ​​of the relevant evaluation results are assigned. Representative Irrelevant evaluation results, For the The total number of irrelevant evaluation results of knowledge graph data retrieval platforms, For the The evaluation assignment is irrelevant to the evaluation results. is the standardized coefficient.

4. The efficient retrieval method for structured data based on knowledge graph according to claim 1 is characterized in that: The comparing of the plurality of recommendation evaluation values ​​and selecting a recommended search platform from the plurality of knowledge graph data search platforms specifically comprises the following steps: Comparing the plurality of recommendation evaluation values ​​to generate a recommendation comparison result; According to the recommendation comparison result, selecting the best evaluation value from the plurality of recommended evaluation values; From the multiple knowledge graph data retrieval platforms, select a recommended retrieval platform corresponding to the best evaluation value.

5. The efficient retrieval method for structured data based on knowledge graph according to claim 1 is characterized in that: The method of performing efficient search processing on the plurality of key search information based on the recommendation search platform and generating and displaying the search result information specifically includes the following steps: Based on the recommendation search platform, a plurality of the key search information are efficiently searched and processed to generate search result information; Create a search display interface; In the search display interface, the search result information is displayed.

6. An efficient retrieval system for structured data based on knowledge graph, characterized in that: The system includes a search request analysis module, a limited data acquisition module, a comprehensive evaluation analysis module, a recommendation platform selection module and an efficient search display module, wherein: A search request analysis module is used to receive a comprehensive search request from a user, analyze the comprehensive search request, obtain a target search requirement and multiple key search information, and determine whether an efficient search is required; A limited data acquisition module is used to obtain limited search evaluation data from multiple preset knowledge graph data retrieval platforms when efficient retrieval is required; A comprehensive evaluation analysis module, used to perform relevant and irrelevant comprehensive evaluation analysis on the plurality of limited search evaluation data, and calculate the recommended evaluation values ​​corresponding to the plurality of knowledge graph data retrieval platforms; Identify multiple search key information, determine the search related types, and then identify the relevant historical evaluations of multiple limited search evaluation data according to the search related types, determine multiple relevant evaluation results with the same search related types in each limited search evaluation data, and at the same time, determine multiple irrelevant evaluation results with different search related types in each limited search evaluation data, and import preset coefficient setting data, and then calculate the recommended evaluation values ​​corresponding to multiple knowledge graph data retrieval platforms according to the coefficient setting data, multiple relevant evaluation results and multiple irrelevant evaluation results; A recommendation platform selection module, used to compare the multiple recommendation evaluation values ​​and select a recommended retrieval platform from the multiple knowledge graph data retrieval platforms; An efficient search and display module, used to perform efficient search processing on a plurality of the key search information based on the recommended search platform, and generate and display search result information; The limited data acquisition module specifically includes: A record acquisition unit, used to acquire historical search record data from multiple preset knowledge graph data retrieval platforms when efficient retrieval is required; An evaluation and screening unit, used for performing evaluation, identification and screening on the plurality of historical search record data to generate a plurality of historical search evaluation data; The limited screening unit is used to screen the latest limited retrieval evaluation data from the multiple historical retrieval evaluation data according to the preset limited quantity requirements.

7. The efficient structured data retrieval system based on knowledge graph according to claim 6 is characterized in that: The retrieval request analysis module specifically includes: A request interaction unit, used to perform request interaction in a preset search request interface and receive a user's comprehensive search request; A request analysis unit, used to analyze the comprehensive search request to obtain search request information and target search requirements; A key identification unit, used to perform key identification on the search request information, and extract multiple search key information from the search request information; The requirement judgment unit is used to identify the target search requirement and judge whether an efficient search is required.

Images