Search

CN-121996758-A - Problem processing method and device based on artificial intelligence, computer equipment and medium

CN121996758ACN 121996758 ACN121996758 ACN 121996758ACN-121996758-A

Abstract

The application belongs to the technical field of artificial intelligence, and relates to a problem processing method, a device, computer equipment and a storage medium based on artificial intelligence, wherein the method comprises the steps of receiving a problem text input by a user; the method comprises the steps of carrying out task analysis on a question text to obtain a task analysis result, retrieving a target strategy unit related to the task analysis result from an experience library, fusing the target strategy unit and the question text to obtain a target prompt sequence, carrying out strategy path selection and combination processing on the target strategy unit based on a graph neural network to obtain a target strategy path, carrying out execution processing on the target strategy path based on the target prompt sequence by using a large language model to generate corresponding answer data, and carrying out output processing on the answer data if the answer data passes knowledge verification. In addition, answer data may be stored in the blockchain. The application can be applied to question-answer processing scenes in the field of finance and technology, effectively improves the processing efficiency of question processing, and improves the accuracy of answer data.

Inventors

  • ZHANG NAN

Assignees

  • 平安科技（深圳）有限公司

Dates

Publication Date
20260508
Application Date
20260106

Claims (10)

  1. 1. The problem processing method based on artificial intelligence is characterized by comprising the following steps: Receiving a question text input by a user; performing task analysis on the problem text based on a preset analysis strategy to obtain a corresponding task analysis result; searching a target strategy unit related to the task analysis result from a preset experience library; Performing fusion processing on the target strategy unit and the problem text to obtain a corresponding target prompt sequence; Selecting and combining the strategy paths of the target strategy units based on a preset graph neural network to obtain corresponding target strategy paths; based on the target prompt sequence, executing processing on the target strategy path by using a preset large language model to generate corresponding answer data; Carrying out knowledge verification on the answer data; and if the answer data passes the knowledge verification, outputting the answer data.
  2. 2. The method for processing the problem based on the artificial intelligence according to claim 1, wherein the step of performing task analysis on the problem text based on a preset analysis policy to obtain a corresponding task analysis result specifically comprises: Preprocessing the problem text to obtain a corresponding target text; performing natural language understanding processing on the target text to obtain a corresponding understanding result; performing matching processing based on the understanding result and a preset task type template to generate a task type corresponding to the target text; comparing and mapping the task type with tasks in a preset historical task database to obtain a corresponding processing result; And taking the processing result as a task analysis result corresponding to the problem text.
  3. 3. The method for processing an artificial intelligence-based problem according to claim 1, wherein the step of retrieving a target policy unit related to the task parsing result from a preset experience library comprises: Calling a preset experience library; Performing similarity calculation on the task analysis result and each strategy unit contained in the experience library to obtain a corresponding similarity calculation result; Based on the similarity calculation result and a preset similarity threshold value, a first strategy unit meeting a similarity condition between the task analysis result and the experience library is retrieved; Screening the first strategy unit based on a preset strategy unit screening strategy to obtain a corresponding second strategy unit; And taking the second strategy unit as the target strategy unit.
  4. 4. The method for processing the problem based on the artificial intelligence according to claim 1, wherein the step of fusing the target policy unit and the problem text to obtain the corresponding target prompt sequence specifically includes: performing functional analysis on the target strategy unit to obtain a corresponding functional analysis result; performing context extraction processing on the problem text to obtain corresponding context information; generating corresponding instruction information based on the function analysis result and the context information; and carrying out organization processing on the instruction information based on a preset organization strategy to generate a corresponding target prompt sequence.
  5. 5. The method for processing the problem based on the artificial intelligence according to claim 1, wherein the step of selecting and combining the policy paths of the target policy unit based on the preset graph neural network to obtain the corresponding target policy paths specifically includes: Constructing a corresponding policy unit relation diagram based on the target policy unit; Performing feature extraction processing on the problem text to obtain a corresponding feature vector; Inputting the feature vector into the strategy unit relation diagram, and carrying out reasoning processing on the strategy unit relation diagram based on a preset graph neural network to obtain score data of each strategy unit node; Selecting and combining corresponding optimal policy paths from the policy unit relation graph based on the score data; and taking the optimal strategy path as the target strategy path.
  6. 6. The method for processing an artificial intelligence based question according to claim 1, wherein the step of performing processing on the target policy path using a preset large language model based on the target prompt sequence to generate corresponding answer data specifically includes: inputting the target prompt sequence and the target strategy path into the large language model; based on the large language model, gradually executing the target strategy path according to the target prompt sequence to obtain a plurality of corresponding strategy processing results; integrating all the strategy processing results based on a preset language generation rule to obtain corresponding integrated data; And taking the integrated data as the answer data.
  7. 7. The method for processing questions based on artificial intelligence as claimed in claim 1, wherein the step of performing knowledge verification on the answer data comprises: calling a preset clause library and a rule library; Performing semantic verification on the answer data based on the clause library; if the answer data passes the semantic verification, carrying out compliance verification on the answer data based on the rule base; And if the answer data passes the compliance verification, judging that the answer data passes the knowledge verification, otherwise, judging that the answer data does not pass the knowledge verification.
  8. 8. An artificial intelligence based problem-handling device, comprising: the receiving module is used for receiving the question text input by the user; The analysis module is used for carrying out task analysis on the problem text based on a preset analysis strategy to obtain a corresponding task analysis result; the retrieval module is used for retrieving a target strategy unit related to the task analysis result from a preset experience library; the fusion module is used for carrying out fusion processing on the target strategy unit and the problem text to obtain a corresponding target prompt sequence; The processing module is used for selecting and combining the strategy paths of the target strategy units based on a preset graph neural network to obtain corresponding target strategy paths; the generation module is used for executing processing on the target strategy path by using a preset large language model based on the target prompt sequence to generate corresponding answer data; the verification module is used for carrying out knowledge verification on the answer data; and the output module is used for outputting the answer data if the answer data passes the knowledge verification.
  9. 9. A computer device comprising a memory having stored therein computer readable instructions which when executed implement the steps of the artificial intelligence based problem handling method of any of claims 1 to 7.
  10. 10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the artificial intelligence based problem handling method according to any of claims 1 to 7.

Description

Problem processing method and device based on artificial intelligence, computer equipment and medium Technical Field The application relates to the technical field of artificial intelligence, which can be applied to the field of financial science and technology, in particular to a problem processing method, a device, computer equipment and a storage medium based on artificial intelligence. Background In the insurance finance field, along with the continuous development of artificial intelligence technology, a Large Language Model (LLMs) has been gradually applied to various problem processing scenes in recent years, and covers multiple aspects of intelligent customer service, insurance clause interpretation, claim settlement assistance, compliance auditing and the like, so that powerful technical support is provided for the development of insurance business. However, there are significant drawbacks to current products that deal with insurance business problems based on large language models in the industry. The existing products generally depend on a large language model and a manually constructed prompt project to carry out problem processing work. Specifically, for different insurance business scenarios, such as health insurance questions and answers, vehicle insurance claims and audits, financial compliance checks, etc., corresponding experience prompts need to be specially designed by experts. The process requires a great deal of time and effort by experts, which not only results in extremely low efficiency of problem processing, but also cannot ensure that the generated answers have high accuracy due to subjectivity and limitation of manual design, and is difficult to meet strict requirements of insurance finance industry on the quality and efficiency of problem processing. Therefore, there is a need to develop a new technology capable of improving the problem processing efficiency of insurance business and ensuring the answer accuracy. Disclosure of Invention The embodiment of the application aims to provide a problem processing method, device, computer equipment and storage medium based on artificial intelligence, so as to solve the technical problems that the existing problem processing mode has low processing efficiency and cannot guarantee the accuracy of answers. In a first aspect, there is provided an artificial intelligence based problem handling method, comprising: Receiving a question text input by a user; performing task analysis on the problem text based on a preset analysis strategy to obtain a corresponding task analysis result; searching a target strategy unit related to the task analysis result from a preset experience library; Performing fusion processing on the target strategy unit and the problem text to obtain a corresponding target prompt sequence; Selecting and combining the strategy paths of the target strategy units based on a preset graph neural network to obtain corresponding target strategy paths; based on the target prompt sequence, executing processing on the target strategy path by using a preset large language model to generate corresponding answer data; Carrying out knowledge verification on the answer data; and if the answer data passes the knowledge verification, outputting the answer data. In a second aspect, there is provided an artificial intelligence based problem handling apparatus comprising: the receiving module is used for receiving the question text input by the user; The analysis module is used for carrying out task analysis on the problem text based on a preset analysis strategy to obtain a corresponding task analysis result; the retrieval module is used for retrieving a target strategy unit related to the task analysis result from a preset experience library; the fusion module is used for carrying out fusion processing on the target strategy unit and the problem text to obtain a corresponding target prompt sequence; The processing module is used for selecting and combining the strategy paths of the target strategy units based on a preset graph neural network to obtain corresponding target strategy paths; the generation module is used for executing processing on the target strategy path by using a preset large language model based on the target prompt sequence to generate corresponding answer data; the verification module is used for carrying out knowledge verification on the answer data; and the output module is used for outputting the answer data if the answer data passes the knowledge verification. In a third aspect, a computer device is provided, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the artificial intelligence based problem handling method described above when the computer program is executed by the processor. In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program whi