CN-121980001-A - Intelligent question-answering and knowledge pushing system and method for carbon platform based on AI analysis

CN121980001ACN 121980001 ACN121980001 ACN 121980001ACN-121980001-A

Abstract

The invention relates to the technical field of carbon management platforms, and discloses an intelligent question-answering and knowledge pushing system and method for a carbon platform based on AI analysis; the intelligent accurate judgment of the carbon-based consultation platform required by the carbon-based consultation is carried out according to the user carbon-based consultation cost requirement information, the user carbon-based consultation problem type identification information and the carbon-based consultation user occupation type identification information and by combining an AI intelligent algorithm with the carbon-based consultation requirement characteristic information of different carbon-based consultation platforms stored on big data, the carbon-based consultation platform is built in a multi-dimensional intelligent depth mode based on the carbon-based consultation problems, the user occupation and the cost, the carbon-based consultation result pushing mode is selected in an individualized mode according to the carbon-based consultation user occupation type identification information and by combining data analysis and scientifically stored information of different carbon-based consultation user occupation carbon-based consultation result pushing modes, the personalized matching of the carbon-based consultation result pushing mode based on the user occupation is realized, and the user experience of carbon-based consultation pushing is improved.

Inventors

HE FANG
CHEN RUIJUN
TANG HUQIANG
ZHANG WEI
LIU YAN

Assignees

红有软件股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260409

Claims (10)

1. The intelligent carbon platform question-answering and knowledge pushing method based on AI analysis is characterized by comprising the following steps: The method comprises the steps of respectively acquiring carbon-capable consultation content information of a user, occupation characteristic information of the carbon-capable consultation user and carbon-capable consultation cost requirement information of the user, carrying out carbon-capable consultation problem type identification processing based on the carbon-capable consultation information to obtain carbon-capable consultation problem type identification information of the user; Judging the carbon question-answering platform required by the carbon consultation to obtain the judging information of the carbon question-answering platform, re-executing the carbon question-answering pushing operation until the carbon question-answering pushing operation exists when the carbon question-answering pushing operation does not exist, and solving the carbon consultation problem when the carbon question-answering pushing operation exists to obtain the carbon consultation result information; selecting the pushing mode of the carbon consultation result to obtain the selecting information of the pushing mode of the carbon consultation result, constructing the pushing information of the carbon consultation result and executing the pushing operation of the carbon consultation result.
2. The intelligent carbon platform question-answering and knowledge pushing method based on AI analysis of claim 1, wherein the method is characterized by respectively collecting carbon consultation content information of a user, occupation characteristic information of the carbon consultation user and carbon consultation cost requirement information of the user, carrying out carbon consultation problem type identification processing based on the carbon consultation information to obtain carbon consultation problem type identification information of the user, carrying out carbon consultation user occupation type identification processing according to the carbon consultation user occupation information, and obtaining carbon consultation user occupation type identification information, and comprises the following operation steps: The method comprises the steps of collecting concrete content text information of a carbon consultation enabling user on line through an information input dialog box, and generating carbon consultation enabling content information of the user; the method comprises the steps of acquiring specific engaging professional characteristic information of a carbon consultation enabling user on line through an information input dialog box, and generating professional characteristic information of the carbon consultation enabling user; performing carbon consultation problem type identification processing based on the user carbon consultation content information and different carbon consultation problem feature keyword sets to obtain user carbon consultation problem type identification information; And carrying out the career type recognition processing of the carbon consultation enabling user according to the career characteristic information of the carbon consultation enabling user and the different career characteristic keyword sets of the carbon consultation enabling user to obtain career type recognition information of the carbon consultation enabling user.
3. The AI analysis-based carbon platform intelligent question-answering and knowledge pushing method as set forth in claim 2, wherein the operation steps of performing carbon consultation question type identification processing based on the user carbon consultation content information and different carbon consultation question feature keyword sets to obtain user carbon consultation question type identification information are as follows: establishing different carbon consultation problem characteristic keyword sets The said Comprising Wherein Represent the first Different carbon consultation problem feature keywords corresponding to the carbon consultation problem types; adopting a breadth-first search algorithm to combine the user carbon consultation content information with the user carbon consultation content information The said in Matching the keyword of the carbon consultation-capable problem, and searching out the keyword matched with the carbon consultation-capable content information of the user Corresponding text information of the carbon consultation question type, and generating identification information of the user carbon consultation question type through data identification.
4. The intelligent carbon platform question-answering and knowledge pushing method based on AI analysis of claim 3, wherein the operation steps of carrying out carbon consultation user occupation type identification processing according to the carbon consultation user occupation characteristic information and different carbon consultation user occupation characteristic keyword sets to obtain carbon consultation user occupation type identification information are as follows: establishing different carbon consultation user occupational characteristic keyword sets The said Comprising Wherein Represent the first Different carbon consultation user occupation characteristic keywords corresponding to the user occupation types; adopting a unified cost search algorithm to combine the occupational characteristic information of the carbon consultation user with the occupation characteristic information The said in Performing occupation keyword matching to search out the occupation characteristic information matched with the carbon consultable user Corresponding user occupation type text information, and generating carbon consultation user occupation type identification information through data identification.
5. The AI analysis-based intelligent carbon platform question-answering and knowledge pushing method as set forth in claim 4, wherein the method comprises the steps of judging a carbon-capable question-answering platform required by carbon consultation to obtain carbon-capable question-answering platform judgment information, re-executing carbon-capable question-answering pushing operation until the carbon-capable question-answering pushing operation exists when the carbon-capable question-answering platform judgment information does not exist, and solving the carbon-capable consultation problem when the carbon-capable question-answering operation is performed to obtain carbon-capable consultation result information, wherein the operation steps are as follows: Performing carbon-based question-answering platform judgment processing required by carbon-based consultation according to the user carbon-based consultation cost requirement information, the user carbon-based consultation problem type identification information, the carbon-based consultation user occupation type identification information and different carbon-based question-answering platform carbon-based consultation requirement characteristic information sets to obtain carbon-based question-answering platform judgment information; and when the carbon-based consultation information exists, carrying out carbon-based consultation problem answering processing according to the carbon-based question answering platform judging information and the carbon-based consultation content information of the user to obtain carbon-based consultation result information.
6. The AI analysis-based carbon platform intelligent question-answering and knowledge pushing method as set forth in claim 5, wherein the carbon-based question-answering platform judgment processing required by carbon consultation is performed according to the user carbon consultation cost requirement information, the user carbon consultation problem type identification information, the carbon consultation user occupation type identification information and different carbon-based question-answering platform carbon consultation requirement characteristic information sets to obtain carbon-based question-answering platform judgment information, and when the carbon-based question-answering platform judgment information does not exist, the operation steps of re-executing the carbon-based question-answering pushing operation until the carbon-based question-answering platform judgment information exists are as follows: Establishing different carbon-based question-answering platforms and carbon-based consultation requirement characteristic information sets The said Comprising Wherein Represent the first Different carbon-based question-answering platforms corresponding to the carbon-based question-answering platform types can be used for carbon consultation requirement characteristic information; the carbon consultation cost requirement information of the user, the carbon consultation problem type identification information of the user, the occupation type identification information of the carbon consultation user and the carbon consultation requirement information of the user The said in Performing carbon consultation requirement text matching, generating carbon question-answering platform judgment information according to a carbon consultation requirement text matching result, and executing the operation steps of generating the carbon question-answering platform judgment information as follows: Step 2121, initializing, at said Initialization in search space of (a) The carbon-enabled question-answering platform search particles initialize the following attributes respectively: First, the The individual carbon question-answering platform searches the position of the particle: , ; First, the Speed of searching particles by the individual carbon question-answering platform: ; First, the The carbon-based question-answering platform searches the optimal position where the particle passes: ; the whole carbon question-answering platform searches the optimal position passed by the particle swarm: ; adding a limit to the positions of all carbon question-answering platform search particles: Wherein And And respectively representing the carbon-based question-answering platform search particles in the different carbon-based question-answering platforms to obtain the feature information set of the carbon-based consultation requirement An upper boundary and a lower boundary of a position in a search space of (a); the speed of searching particles by all carbon question-answering platforms is limited: Wherein And And respectively representing the carbon-based question-answering platform search particles in the different carbon-based question-answering platforms to obtain the feature information set of the carbon-based consultation requirement An upper speed boundary and a lower speed boundary in a search space of (a); setting the maximum iteration number ; During each iteration, the following is set The carbon question-answering platform can search the self-learning factors of the particles in the search space; during each iteration, the following is set The carbon question-answering platform can search the group learning factors of the particles in the search space; during each iteration, the following is set The carbon question-answering platform can search the inertial weight of the particles in the search space; step 2122, calculating an adaptability value of the carbon question-answering platform search particles: At the said Calculate the first in the search space of (2) Searching the particles by a carbon question-answering platform The adaptability value of the carbon consultation request information, the carbon consultation question type identification information and the occupation type identification information of the carbon consultation user; Step 2123, updating a carbon question-answering platform to search individual extremum and a global optimal solution of particles: At the said Update in search space of (c) Searching the optimal fitness value of the particle individual by the personal carbon question-answering platform And a carbon question-answering platform searching for an optimal fitness value of the population of particles as a whole According to At the said Updating optimal positions of search particles of carbon question-answering platform in search space of (a) From these optimal positions Searching out the optimal position of a group, and combining the optimal position Marking as the global optimal position of the iteration; Step 2124, updating the carbon question-answering platform to search individual particles in the question Speed and position in the search space of (a); If in the iterative process, the first Searching the position of particles by a carbon question-answering platform Exceeding the boundary The carbon question-answering platform searches for particles Is adjusted to Or (b) First, a third step Speed of particle search by carbon-based question-answering platform Exceeding the boundary The carbon question-answering platform searches for particles Is adjusted to Or (b) ; Step 2125, setting termination condition, when setting maximum iteration number When the user can carbon consultation cost requirement information, the user can carbon consultation problem type identification information, the carbon consultation user occupation type identification information and the user can carbon consultation request information are used for identifying the user occupation type The said in The carbon consultation requirement text matching result generates carbon question-answering platform judgment information; When the user can carbon consultation cost requirement information, the user can carbon consultation problem type identification information, the carbon consultation user occupation type identification information and the user can carbon consultation request information When the matching of the carbon consultation requirement text is successful, outputting the judging information of the carbon consultation enabling platform to be present, and outputting the information which is matched with the carbon consultation enabling cost requirement information of the user, the carbon consultation enabling problem type identification information of the user and the occupation type identification information of the carbon consultation enabling user Corresponding text information of the carbon question-answering platform type; When the user can carbon consultation cost requirement information, the user can carbon consultation problem type identification information, the carbon consultation user occupation type identification information and the user can carbon consultation request information And when the carbon consultation request texts are successfully matched, outputting the judging information of the carbon question-answering platform, and re-executing the carbon question-answering pushing operation until the judging information of the carbon question-answering platform is present.
7. The intelligent carbon platform question-answering and knowledge pushing method based on AI analysis of claim 6, wherein when the intelligent carbon platform question-answering and knowledge pushing method based on AI analysis exists, the method is characterized in that according to the judging information of the carbon question-answering platform and the carbon consultation content information of the user, the carbon consultation question-answering processing is carried out, and the operation steps of obtaining the carbon consultation result information are as follows: When the carbon-based question-answering platform judging information is present, the carbon-based consultation content information of the user is input into an information input dialog box of a carbon-based question-answering platform corresponding to the carbon-based question-answering platform judging information, and the carbon-based question-answering platform retrieves carbon-based consultation answering text information required by the carbon-based consultation user and generates carbon-based consultation result information.
8. The AI analysis-based intelligent question-answering and knowledge pushing method of a carbon platform of claim 7, wherein the method comprises the steps of selecting a carbon consultation result pushing mode to obtain carbon consultation result pushing mode selection information, constructing carbon consultation result pushing information and executing carbon consultation result pushing operation, and comprises the following steps: performing carbon-based consultation result pushing mode selection processing according to the carbon-based consultation user occupation type identification information and different carbon-based consultation user occupation carbon-based consultation result pushing mode information sets to obtain carbon-based consultation result pushing mode selection information; Constructing carbon consultation result pushing information by data combination identification of the carbon consultation result information and the carbon consultation result pushing mode selection information; and feeding back the carbon-based consultation result information to the user according to the selected carbon-based consultation result pushing method through the carbon-based question and answer platform based on the carbon-based consultation result pushing information, and executing the carbon-based consultation result pushing operation.
9. The intelligent carbon platform question-answering and knowledge pushing method based on AI analysis of claim 8, wherein the operation steps of selecting and processing the carbon consultation result pushing mode according to the carbon consultation user occupation type identification information and different carbon consultation user occupation carbon consultation result pushing mode information sets to obtain carbon consultation result pushing mode selection information are as follows: establishing different carbon consultation user occupational carbon consultation result pushing mode information set The said Comprising Wherein Represent the first Different carbon consultation user occupation carbon consultation result pushing mode information corresponding to the user occupation types; The career type identification information of the carbon consultable user is recorded and the The said in And matching the keyword of the user occupation type, searching out information of pushing modes of the carbon consultation results of different carbon consultation users corresponding to the identification information of the carbon consultation user occupation type, and generating selection information of pushing modes of the carbon consultation results through data identification.
10. The intelligent carbon platform question-answering and knowledge pushing system based on the AI analysis is used for realizing the intelligent carbon platform question-answering and knowledge pushing method based on the AI analysis, and is characterized by comprising a carbon consultation question preprocessing module, a carbon consultation question answering module and a carbon consultation result pushing module.

Description

Intelligent question-answering and knowledge pushing system and method for carbon platform based on AI analysis Technical Field The invention relates to the technical field of carbon-capable management platforms, in particular to an intelligent question-answering and knowledge pushing system and method for a carbon-capable platform based on AI analysis. Background The carbon question-answering pushing platform is an intelligent system based on deep fusion, and has the technical architecture core that an intelligent engine in the field is constructed through a knowledge graph and a large language model, unstructured documents such as scattered policy regulations, accounting standards, industry cases and the like are subjected to vectorization processing to form a professional knowledge base capable of being subjected to semantic retrieval, the platform is integrated with an Internet of things and an enterprise information system to realize real-time acquisition and management of energy consumption and carbon emission data, dynamic data support is provided for question-answering, user intention is understood by using a natural language processing technology in an interaction layer, accurate answers or recommendation association analysis reports are generated by driving the large model, the traditional carbon question-answering pushing platform cannot realize the intelligent answer of users based on multidimensional information and cannot provide personalized carbon consultation result pushing services, and user experience and intelligence of the carbon question-answering platform are reduced. The Chinese patent invention with the bulletin number of CN119129942B and the bulletin date of 2025.02.14 discloses a campus carbon metering method, an intelligent question-answering method and a device based on a knowledge graph, carbon activity attribute data and time attribute data of initial main body attribute nodes are obtained according to the initial knowledge graph by constructing the initial knowledge graph of carbon activity content in the campus, the time attribute data corresponds to the time nodes, carbon emission calculation tasks are executed according to the carbon activity attribute data and the time attribute data of the initial main body attribute nodes and carbon emission calculation parameters related to the carbon activity nodes to obtain main body carbon emission information related to the initial main body attribute nodes, however, the technical scheme cannot realize intelligent answer of carbon consultation requirements of users based on multidimensional information, and personalized carbon consultation result pushing service cannot be provided. Disclosure of Invention (One) solving the technical problems In order to solve the problems that the conventional carbon-enabled question-answering pushing platform cannot realize the carbon-enabled consultation requirement of the user based on multidimensional information and cannot provide personalized carbon-enabled consultation result pushing service, the user experience and the intelligent problem of the carbon-enabled question-answering platform are reduced, the purposes of respectively acquiring the carbon-enabled consultation content information of the user, the occupation characteristic information of the carbon-enabled consultation user and the carbon-enabled consultation cost requirement information of the user, accurately identifying the carbon-enabled consultation problem type and the occupation type of the carbon-enabled consultation user, accurately judging the carbon-enabled question-answering pushing platform based on the carbon-enabled consultation problem type, the user occupation type and the cost multidimensional precision, accurately solving the carbon-enabled consultation problem of the user, personally pushing the carbon-enabled consultation result of the user and improving the user experience and the intelligent purpose of the carbon-enabled question-answering pushing platform are achieved. (II) technical scheme The invention is realized by the following technical scheme that the carbon platform intelligent question-answering and knowledge pushing method based on AI analysis comprises the following steps: The method comprises the steps of respectively acquiring carbon-capable consultation content information of a user, occupation characteristic information of the carbon-capable consultation user and carbon-capable consultation cost requirement information of the user, carrying out carbon-capable consultation problem type identification processing based on the carbon-capable consultation information to obtain carbon-capable consultation problem type identification information of the user; Judging the carbon question-answering platform required by the carbon consultation to obtain the judging information of the carbon question-answering platform, re-executing the carbon question-answering pushing operation until the carbon question