Resource data comprehensive query method and system based on dynamic cue word and Text2Api

Abstract

The invention relates to the technical field of computer information, in particular to a resource data comprehensive query method and system based on a dynamic cue word and Text2Api, and the method comprises the following steps: carrying out the preliminary analysis of a conversation sentence input by a user through employing an Elasticsearch technology, constructing a detailed dynamic cue word, and carrying out the calculation of the dynamic cue word; sending the constructed dynamic cue word and the original session content of the user to a large language model (LLM) for deep analysis, analyzing and determining the name and the parameter of an API which actually needs to be called, and acquiring data required by the user from a corresponding data source; the resource management system has the beneficial effects that various resource data query experiences are provided for the user at a unified entry, the user can effectively obtain the resource data which the user wants to query on the basis that the user is not familiar with the resource management system, the operation logic is simplified, and the user operation experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer information technology, and specifically provides a comprehensive resource data query method and system based on dynamic prompt words and Text2Api. Background Art

[0002] With the development of technology and the upgrade of resource management systems, the types of resource data managed by operators are increasing, and the demand for data query is also growing. Traditional data query methods often require users to be familiar with the system and have certain professional knowledge to correctly construct query statements to obtain the required information.

[0003] In addition, due to the diverse data sources, there are differences in API interfaces between different data sources, which poses challenges to cross-platform data retrieval.

[0004] Therefore, how to help users quickly and accurately obtain the required data in a more intelligent way has become an urgent problem to be solved. Summary of the Invention

[0005] The purpose of the present invention is to provide a comprehensive resource data query method and system based on dynamic prompt words and Text2Api, which combines dynamic generation of prompt words and Text2Api technology to achieve effective access to various resource data, and can provide users with a query experience of various resource data at a unified entrance, so as to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solution: A comprehensive resource data query method based on dynamic prompt words and Text2Api, the method includes the following steps:

[0007] a) Use Elasticsearch technology to preliminarily analyze the conversation sentence input by the user, identify the basic topic classification to which the conversation sentence belongs and the possible data model types involved, and the topic classification includes resource query, topology query, statistical analysis, work order query, coverage ability query;

[0008] b) According to the topic classification and data model type identified in step a), search and determine the corresponding detailed data or classification information in the established comprehensive resource library, and construct detailed dynamic prompt words accordingly;

[0009] c) Send the constructed dynamic prompt words and the original conversation content of the user to the large language model LLM for in-depth parsing to further understand the user's query intention;

[0010] d) The large language model LLM analyzes and determines the actual API name and its parameters to be called according to the received information;

[0011] e) Execute a data request operation using the API name and parameters determined in step d) to obtain the data required by the user from the corresponding data source.

[0012] Preferably, the process of constructing dynamic prompt words in step b) further includes:

[0013] If it is identified as a topology - type problem, query and construct basic topology classification prompt words;

[0014] If it is identified as a work order - type problem, query and construct existing work order classification prompt words;

[0015] If it is identified as a free query, query and construct corresponding model attribute information prompt words.

[0016] Preferably, the method further includes:

[0017] f) According to the type of the API and user preferences, display the query results to the user in various ways such as text and graphics to enhance the user's data understanding and operation experience.

[0018] Preferably, Elasticsearch technology is used to achieve efficient indexing and fast retrieval of user conversation sentences, improving the accuracy and efficiency of topic classification and data model type recognition.

[0019] Preferably, the large - language model LLM, through deep learning and natural language processing technologies, deeply understands and analyzes the conversation content and dynamic prompt words input by the user, accurately parses the user's query intention, and generates accurate API call instructions to achieve effective access and query of various resource data.

[0020] A comprehensive resource data query system based on dynamic prompt words and Text2Api is applied to a comprehensive resource data query method based on dynamic prompt words and Text2Api. The system includes:

[0021] A pre - analysis module that uses Elasticsearch technology to preliminarily analyze the conversation sentences input by the user, identify the basic topic classification to which the conversation sentences belong and the possible data model types involved. The topic classification includes resource query, topology query, statistical analysis, work order query, and coverage ability query;

[0022] A dynamic prompt word construction module that, according to the topic classification and data model types identified by the analysis module, searches and determines the corresponding detailed data or classification information in the established comprehensive resource library, and constructs detailed dynamic prompt words based on this;

[0023] The parsing module sends the constructed dynamic prompt words and the user's original conversation content to the large language model (LLM) for in-depth parsing to further understand the user's query intention;

[0024] The analysis module. The large language model (LLM) analyzes and determines the actual API names and their parameters to be called based on the received information;

[0025] The data acquisition module performs a data request operation using the API names and parameters determined in the step analysis module to obtain the data required by the user from the corresponding data source.

[0026] Preferably, the dynamic prompt word construction module further includes:

[0027] If it is identified as a topology class problem, query and construct basic topology classification prompt words;

[0028] If it is identified as a work order class problem, query and construct existing work order classification prompt words;

[0029] If it is identified as a free query, query and construct corresponding model attribute information prompt words.

[0030] Preferably, the system further includes: a result display module: according to the type of the API and the user's preference, presents the query results to the user in various ways such as text and graphics, improving the user's data understanding and operation experience.

[0031] Preferably, in the pre-analysis module, the Elasticsearch technology is used to achieve efficient indexing and rapid retrieval of the user's conversation sentences, improving the accuracy and efficiency of topic classification and data model type recognition.

[0032] Preferably, in the analysis module, the large language model (LLM) deeply understands and analyzes the user's input conversation content and dynamic prompt words through deep learning and natural language processing technologies, accurately parses the user's query intention, and generates accurate API call instructions to achieve effective access and query of various resource data.

[0033] Compared with the prior art, the beneficial effects of the present invention are:

[0034] The resource data comprehensive query method and system based on dynamic prompt words and Text2Api proposed by the present invention provide the user with a query experience of various resource data at a unified entrance, enabling the user to effectively obtain the resource data they want to query even without being familiar with the resource management system, simplifying the operation logic, and enhancing the user's operation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic flow chart of the method of the present invention. Detailed implementation manners

[0036] In order to clearly and completely describe the objectives and technical solutions of the present invention, and make the advantages more clearly understood, the following further details the embodiments of the present invention with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are part of the embodiments of the present invention, rather than all of the embodiments, and are only used to explain the embodiments of the present invention, not to limit the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0037] Example 1. Please refer to Figure 1 , the present invention provides a technical solution: a comprehensive resource data query method based on dynamic prompt words and Text2Api, and the method includes the following steps:

[0038] a) Use Elasticsearch technology to preliminarily analyze the conversation sentences input by the user, identify the basic topic classification to which the conversation sentences belong and the possible data model types involved. The topic classification includes resource query, topology query, statistical analysis, work order query, and coverage ability query; Elasticsearch technology is used to achieve efficient indexing and fast retrieval of the user's conversation sentences, improving the accuracy and efficiency of topic classification and data model type identification.

[0039] b) According to the topic classification and data model types identified in step a), search and determine the corresponding detailed data or classification information in the established comprehensive resource library, and construct detailed dynamic prompt words based on this; the process of constructing dynamic prompt words also includes:

[0040] If it is identified as a topology-related problem, query and construct basic topology classification prompt words;

[0041] If it is identified as a work order-related problem, query and construct existing work order classification prompt words;

[0042] If it is identified as a free query, query and construct corresponding model attribute information prompt words.

[0043] c) Send the constructed dynamic prompt words and the user's original conversation content to the large language model LLM for in-depth parsing to further understand the user's query intention.

[0044] d) The large language model LLM analyzes and determines the actual API names and their parameters to be called according to the received information; the large language model LLM uses deep learning and natural language processing technologies to deeply understand and analyze the user's input conversation content and dynamic prompt words, accurately parse the user's query intention, and generate accurate API call instructions to achieve effective access and query of various resource data.

[0045] e) Perform a data request operation using the API name and parameters determined in step d) to obtain the data required by the user from the corresponding data source.

[0046] f) According to the type of API and user preferences, display the query results to the user in various ways such as text and graphics to enhance the user's data understanding and operation experience.

[0047] Example 2, based on Example 1, proposes a comprehensive resource data query system based on dynamic prompt words and Text2Api, which is applied to a comprehensive resource data query method based on dynamic prompt words and Text2Api. The system includes:

[0048] A pre-analysis module uses Elasticsearch technology to preliminarily analyze the conversation sentences input by the user, identify the basic topic classification to which the conversation sentences belong and the possible data model types involved. The topic classification includes resource query, topology query, statistical analysis, work order query, and coverage ability query. Elasticsearch technology is used to achieve efficient indexing and rapid retrieval of the user's conversation sentences, improving the accuracy and efficiency of topic classification and data model type recognition.

[0049] A dynamic prompt word construction module, according to the topic classification and data model type identified in the analysis module, searches and determines the corresponding detailed data or classification information in the established comprehensive resource library, and constructs detailed dynamic prompt words accordingly; it also includes:

[0050] If it is identified as a topology-related problem, query and construct basic topology classification prompt words;

[0051] If it is identified as a work order-related problem, query and construct existing work order classification prompt words;

[0052] If it is identified as a free query, query and construct corresponding model attribute information prompt words.

[0053] A parsing module sends the constructed dynamic prompt words and the user's original conversation content to the large language model LLM for in-depth parsing to further understand the user's query intention;

[0054] An analysis module, based on the information received by the large language model LLM, analyzes and determines the actual API name and its parameters to be called. The large language model LLM, through deep learning and natural language processing technologies, deeply understands and analyzes the user's input conversation content and dynamic prompt words, accurately parses the user's query intention, and generates accurate API call instructions to achieve effective access and query of various resource data.

[0055] The data acquisition module performs a data request operation using the API names and parameters determined in the step analysis module to obtain the data required by the user from the corresponding data source.

[0056] The system further includes a result display module: according to the type of API and user preferences, the query results are displayed to the user in various ways such as text and graphics, improving the user's data understanding and operation experience.

[0057] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A resource data comprehensive query method based on dynamic prompt words and Text2Api, characterized in that: The method comprises the following steps: a) Use Elasticsearch technology to conduct a preliminary analysis of the conversation sentences input by the user, identify the basic topic categories to which the conversation sentences belong and the types of data models that may be involved. The topic categories include resource query, topology query, statistical analysis, work order query, and coverage capability query; b) according to the topic classification and data model type identified in step a), searching and determining the corresponding segmented data or classification information in the established comprehensive resource library, and constructing detailed dynamic prompt words accordingly; c) Send the constructed dynamic prompt words and the user's original conversation content to the large language model LLM for in-depth analysis to further understand the user's query intention; d) The large language model LLM analyzes and determines the API name and parameters that actually need to be called based on the received information; e) Use the API name and parameters determined in step d) to perform data request operations and obtain the data required by the user from the corresponding data source.

2. A resource data comprehensive query method based on dynamic prompt words and Text2Api according to claim 1, characterized in that: The process of constructing the dynamic prompt word in step b) also includes: If it is identified as a topological problem, query and construct basic topological classification prompt words; If it is identified as a work order problem, query and construct the existing work order classification prompt words; If it is identified as a free query, the corresponding model attribute information prompt words are queried and constructed.

3. According to claim 1, a resource data comprehensive query method based on dynamic prompt words and Text2Api is characterized in that: The method also includes: f) Based on the API type and user preferences, query results are displayed to users in a variety of ways, such as text and graphics, to improve users’ data understanding and operational experience.

4. The resource data comprehensive query method based on dynamic prompt words and Text2Api according to claim 1 is characterized in that: Elasticsearch technology is used to achieve efficient indexing and fast retrieval of user conversation sentences, improving the accuracy and efficiency of topic classification and data model type identification.

5. The resource data comprehensive query method based on dynamic prompt words and Text2Api according to claim 1 is characterized in that: The Large Language Model (LLM) uses deep learning and natural language processing technology to deeply understand and analyze the conversation content and dynamic prompt words entered by users, accurately parse the user's query intentions, and generate accurate API call instructions to achieve effective access and query of various resource data.

6. A resource data comprehensive query system based on dynamic prompt words and Text2Api, applied to a resource data comprehensive query method based on dynamic prompt words and Text2Api as described in any one of claims 1 to 5, characterized in that: The system comprises: The pre-analysis module uses Elasticsearch technology to perform preliminary analysis on the conversation sentences input by the user, identifying the basic topic categories to which the conversation sentences belong and the types of data models that may be involved. The topic categories include resource query, topology query, statistical analysis, work order query, and coverage capability query. The dynamic prompt word construction module searches and determines the corresponding segmented data or classification information in the established comprehensive resource library according to the topic classification and data model type identified in the analysis module, and constructs detailed dynamic prompt words accordingly; The parsing module sends the constructed dynamic prompt words and the user's original conversation content to the large language model LLM for in-depth analysis to further understand the user's query intention; The analysis module, the large language model LLM, analyzes and determines the API name and parameters that actually need to be called based on the received information; The data acquisition module uses the API name and parameters determined in the step analysis module to perform data request operations and obtain the data required by the user from the corresponding data source.

7. A resource data comprehensive query system based on dynamic prompt words and Text2Api according to claim 6, characterized in that: The dynamic prompt word building module also includes: If it is identified as a topological problem, query and construct basic topological classification prompt words; If it is identified as a work order problem, query and construct the existing work order classification prompt words; If it is identified as a free query, the corresponding model attribute information prompt words are queried and constructed.

8. A resource data comprehensive query system based on dynamic prompt words and Text2Api according to claim 6, characterized in that: The system also includes: a result display module: according to the type of API and user preferences, the query results are displayed to the user in multiple ways such as text and graphics to improve the user's data understanding and operation experience.

9. A resource data comprehensive query system based on dynamic prompt words and Text2Api according to claim 6, characterized in that: In the pre-analysis module, Elasticsearch technology is used to achieve efficient indexing and fast retrieval of user conversation sentences, improving the accuracy and efficiency of topic classification and data model type identification.

10. The resource data comprehensive query system based on dynamic prompt words and Text2Api according to claim 6, characterized in that: In the analysis module, the large language model LLM uses deep learning and natural language processing technology to deeply understand and analyze the conversation content and dynamic prompt words entered by the user, accurately parse the user's query intention, and generate accurate API call instructions to achieve effective access and query of various resource data.

Images