Question answering list generation method and device, electronic equipment and storage medium

By semantically vectorizing and clustering historical question-and-answer records, a question association network is constructed, and a FAQ list is automatically generated. This solves the problem that traditional manual FAQ list construction cannot adapt to diverse expressions, achieves accurate matching and recall of user questions, and improves user experience and platform efficiency.

CN122364221APending Publication Date: 2026-07-10GUANGZHOU FANGZHOU INFORMATION TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GUANGZHOU FANGZHOU INFORMATION TECHNOLOGY CO LTD
Filing Date
2026-04-07
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Traditional medical and pharmaceutical e-commerce platforms use manual compilation and fixed templates for their FAQ lists, which makes it difficult to cover users' everyday and colloquial expressions. This results in users' questions not being accurately identified and retrieved, thus reducing the user experience.

Method used

By acquiring historical question-and-answer records, performing semantic vectorization and clustering, standard questions are generated, and a question association network is constructed to automatically generate a FAQ list that covers user questions with diverse expressions.

Benefits of technology

It enables precise matching and recall of diverse user questions, improves the timeliness and accuracy of the FAQ list, reduces manual maintenance costs, and enhances user experience and platform service efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122364221A_ABST
    Figure CN122364221A_ABST
Patent Text Reader

Abstract

This invention discloses a method, apparatus, electronic device, and storage medium for generating a question-and-answer list. By vectorizing and clustering the original questions from historical question-and-answer records into multiple categories, a standard question is determined for each cluster. A question association network containing the standard question and the original questions is constructed. A basic question-and-answer list is generated and displayed using the standard question, the original questions, and the answers. When a user selects or inputs a question, a recommended question list is generated and displayed based on the question association network. This achieves automated generation of the question-and-answer list, effectively covering diverse user queries, improving the accuracy of question retrieval and user experience, reducing manual maintenance costs, and solving the problem that manually constructed question-and-answer lists cannot adapt to diverse user expressions, resulting in the inability to provide answers to user questions and a reduced user experience. This allows users to obtain answers more quickly and conveniently, significantly improving user experience and platform service efficiency.
Need to check novelty before this filing date? Find Prior Art