{"title":"基于大语言模型的保险理赔文本分析","authors":"Dongchen Li, Zhuo Jin, Linyi Qian, Hailiang Yang","doi":"10.1111/jori.70004","DOIUrl":null,"url":null,"abstract":"<p>This study proposes a comprehensive and general framework for examining discrepancies in textual content using large language models (LLMs), broadening application scenarios in the insurtech and risk management fields, and conducting empirical research based on actual needs and real-world data. Our framework integrates OpenAI's interface to embed texts and project them into external categories while utilizing distance metrics to evaluate discrepancies. To identify significant disparities, we design prompts to analyze three types of relationships: identical information, logical relationships and potential relationships. Our empirical analysis shows that 22.1% of samples exhibit substantial semantic discrepancies, and 38.1% of the samples with significant differences contain at least one of the identified relationships. The average processing time for each sample does not exceed 4 s, and all processes can be adjusted based on actual needs. Backtesting results and comparisons with traditional NLP methods further demonstrate that our proposed method is both effective and robust.</p>","PeriodicalId":51440,"journal":{"name":"Journal of Risk and Insurance","volume":"92 2","pages":"505-535"},"PeriodicalIF":2.1000,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Textual analysis of insurance claims with large language models\",\"authors\":\"Dongchen Li, Zhuo Jin, Linyi Qian, Hailiang Yang\",\"doi\":\"10.1111/jori.70004\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>This study proposes a comprehensive and general framework for examining discrepancies in textual content using large language models (LLMs), broadening application scenarios in the insurtech and risk management fields, and conducting empirical research based on actual needs and real-world data. Our framework integrates OpenAI's interface to embed texts and project them into external categories while utilizing distance metrics to evaluate discrepancies. To identify significant disparities, we design prompts to analyze three types of relationships: identical information, logical relationships and potential relationships. Our empirical analysis shows that 22.1% of samples exhibit substantial semantic discrepancies, and 38.1% of the samples with significant differences contain at least one of the identified relationships. The average processing time for each sample does not exceed 4 s, and all processes can be adjusted based on actual needs. Backtesting results and comparisons with traditional NLP methods further demonstrate that our proposed method is both effective and robust.</p>\",\"PeriodicalId\":51440,\"journal\":{\"name\":\"Journal of Risk and Insurance\",\"volume\":\"92 2\",\"pages\":\"505-535\"},\"PeriodicalIF\":2.1000,\"publicationDate\":\"2025-03-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Risk and Insurance\",\"FirstCategoryId\":\"96\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1111/jori.70004\",\"RegionNum\":3,\"RegionCategory\":\"经济学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"BUSINESS, FINANCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Risk and Insurance","FirstCategoryId":"96","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/jori.70004","RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"BUSINESS, FINANCE","Score":null,"Total":0}
Textual analysis of insurance claims with large language models
This study proposes a comprehensive and general framework for examining discrepancies in textual content using large language models (LLMs), broadening application scenarios in the insurtech and risk management fields, and conducting empirical research based on actual needs and real-world data. Our framework integrates OpenAI's interface to embed texts and project them into external categories while utilizing distance metrics to evaluate discrepancies. To identify significant disparities, we design prompts to analyze three types of relationships: identical information, logical relationships and potential relationships. Our empirical analysis shows that 22.1% of samples exhibit substantial semantic discrepancies, and 38.1% of the samples with significant differences contain at least one of the identified relationships. The average processing time for each sample does not exceed 4 s, and all processes can be adjusted based on actual needs. Backtesting results and comparisons with traditional NLP methods further demonstrate that our proposed method is both effective and robust.
期刊介绍:
The Journal of Risk and Insurance (JRI) is the premier outlet for theoretical and empirical research on the topics of insurance economics and risk management. Research in the JRI informs practice, policy-making, and regulation in insurance markets as well as corporate and household risk management. JRI is the flagship journal for the American Risk and Insurance Association, and is currently indexed by the American Economic Association’s Economic Literature Index, RePEc, the Social Sciences Citation Index, and others. Issues of the Journal of Risk and Insurance, from volume one to volume 82 (2015), are available online through JSTOR . Recent issues of JRI are available through Wiley Online Library. In addition to the research areas of traditional strength for the JRI, the editorial team highlights below specific areas for special focus in the near term, due to their current relevance for the field.