Problem classification method, device, apparatus, and medium
By acquiring both textual and non-textual information about the issues, and utilizing large language models and log analysis, we can perform two-way interactive issue classification. This solves the problem of inaccurate textual descriptions in user feedback, improves the accuracy of issue classification and analysis efficiency, and enhances the quality of the app.
CN122432341APending Publication Date: 2026-07-21BEIJING QIYI CENTURY SCI & TECH CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING QIYI CENTURY SCI & TECH CO LTD
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-21
AI Technical Summary
Technical Problem
The textual descriptions in user feedback are often insufficient to accurately describe the problems, leading to inaccurate problem categorization and affecting the quality of problem localization and repair for the app.
Method used
A bidirectional interactive problem classification method is developed by acquiring both textual and non-textual information about the problem description, utilizing a large language model and log analysis. This method includes bidirectional interactive processing of the problem description text and log content by the large language model.
Benefits of technology
It improved the accuracy of problem classification, reduced invalid or erroneous analysis, and enhanced the efficiency of problem analysis and the quality of the app.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122432341A_ABST
Abstract
The present disclosure relates to a problem classification method, device, equipment and medium, wherein the method comprises: obtaining feedback information for a problem; the feedback information comprises a problem description text, non-text information and a log; determining a first problem category of the feedback information according to the problem description text and the non-text information; determining a first associated log associated with the first problem category from the log; inputting a large language model according to the first associated log and the problem description text to obtain a second problem category of the feedback information. According to the technical solution of the present disclosure, the accuracy of problem classification can be improved, and invalid or incorrect analysis caused by inaccurate classification during problem analysis is reduced.
Need to check novelty before this filing date? Find Prior Art