基于大语言模型的汽车故障记录自动标注方法研究

打开文本图片集

中图分类号：TP391.1 文献标识码：A文章编号：2096-4706（2025）12-0092-05

Research on an Automatic Annotation Method of Automotive Fault Records Based on Large Language Models

DAI Yawen， ZHAO Yue，ZHANGLin （Wuhu InstituteofTechnology，Wuhu 241ooo， China）

Abstract： Automotive fault records are the crucial data source for automotive condition analysis.However， the unstructured nature and technical specificitymake thecost of manual annotationhigh.Toaddress this isue，this paperproposes an automatic annotation method based onLargeLanguage Models.The studycombines LLMs with domain-specific knowledge byconstructingspecialprompttemplates toextractandanotatekeydata，suchasfaultyparts，faultstatus，anddagnostictools. Experimentalresultsshow thattheDepSek model，using aFew-shot Prompting templatecombined with industryexamples， achieves the best performance，with an annotation precision of 79.65% ，recall of 84.92% ，and F1 score of 82.20% .This outperforms traditional Deep Learning-based annotation methods bymore than 3% in terms ofF1 score.

Keywords： Large language Model; automotive fault record; automatic annotation; Prompt Engineering

0 引言

车辆故障诊断是汽车维修与质量控制的关键环节，在诊断过程中，维修人员会详细记录故障信息。（剩余7462字）