南开保险精算大讲堂是南开保险与精算研究院主办的系列讲座，旨在搭建保险精算领域学术交流平台、推动相关领域的研究与合作。讲座主题涵盖保险精算、风险管理、随机过程、大数据等多个领域，欢迎校内外师生积极参加！

本期南开保险精算大讲堂活动安排如下：

讲座题目

基于大型语言模型的保险理赔文本分析

主讲人：金卓
‍‍‍

Dr. Zhuo Jin is a Professor of Actuarial Studies in the Department of Actuarial Studies and Business Analytics at Macquarie Business School. He serves as the Research Director of the department and is also a Co-Director of the Centre for Emerging Risks at Macquarie University. His research focuses on actuarial science, mathematical finance, risk management, and machine learning. An Associate of the Society of Actuaries (ASA), his research has been supported by the Hong Kong Research Grants Council (RGC) and other funding bodies. He has authored over 70 publications in international journals and book chapters, including top-tier journals such as Insurance: Mathematics and Economics, European Journal of Operational Research, ASTIN Bulletin, Journal of Risk and Insurance, SIAM Journal on Control and Optimization, and Automatica. Additionally, he is a Series Editor for the book series Advances in Statistics, Probability, and Actuarial Science (ASPAS), published by World Scientific.

讲座时间

2025年4月23日（周三）

16：30

讲座地点

金融学院 434

Abstract: 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 seconds, 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.

微信扫一扫
关注该公众号

知道了

微信扫一扫
使用小程序

取消 允许

取消 允许

取消 允许

× 分析

微信扫一扫可打开此内容，
使用完整服务

： ， ， ， ， ， ， ， ， ， ， ， ， 。 视频 小程序 赞 ，轻点两下取消赞 在看 ，轻点两下取消在看 分享 留言 收藏 听过