A Dialogue-based Information Extraction System for Medical Insurance Assessment

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 654–663 August 1–6, 2021. ©2021 Association for Computational Linguistics 654 A Dialogue-based Information Extraction System for Medical Insurance Assessment Shuang Peng∗ Mengdi Zhou∗ Minghui Yang Haitao Mi Shaosheng Cao Zujie Wen Teng Xu Hongbin Wang Lei Liu Ant Group Hangzhou, China {jianfeng.ps,jacquelyn.zmd,minghui.ymh,haitao.mi,shaosheng.css, zujie.wzj,harvey.xt,hongbin.whb,leihu.ll}@antgroup.com Abstract In the Chinese medical insurance industry, the assessor’s role is essential and requires sig- nificant efforts to converse with the claimant. This is a highly professional job that involves many parts, such as identifying personal infor- mation, collecting related evidence, and mak- ing a final insurance report. Due to the coron- avirus (COVID-19) pandemic, the previous of- fline insurance assessment has to be conducted online. However, for the junior assessor of- ten lacking practical experience, it is not easy to quickly handle such a complex online pro- cedure, yet this is important as the insurance company needs to decide how much compen- sation the claimant should receive based on the assessor’s feedback. In order to promote assessors’ work efficiency and speed up the overall procedure, in this paper, we propose a dialogue-based information extraction system that integrates advanced NLP technologies for medical insurance assessment. With the assis- tance of our system, the average time cost of the procedure is reduced from 55 minutes to 35 minutes, and the total human resources cost is saved 30% compared with the previous offline procedure. Until now, the system has already served thousands of online claim cases. 1 Introduction In the Chinese medical insurance industry, the as- sessor’s role is essential in handling protection claims. The insurance assessor may work for the in- surance company or be a third-party administrator assessor. The claimant approaches the insurance assessor to evaluate the disease that has affected them. For example, if the claimant has lung cancer, the insurance assessor would check the medical cost and decide if they are claimable. Once the assessor has finished collecting the evidence and related claimant information (i.e., disease history, ∗Equal contributions. medication experience, in-hospitalized information, etc.), a final detailed case report will be made and sent to the insurance company. The traditional insurance assessment procedure is carried out offline where the assessor conducts an onsite interview with the claimant. Due to the coro- navirus (COVID-19) pandemic, the previous offline insurance assessment has to be interrupted tem- porarily, and thus many claimants can not get their insurance compensation in time. Some of them, meanwhile, may also suffer job loss1. To help these claimants, many insurance companies shift the of- fline procedure to the online platform in which in- surance assessors can communicate with claimants via video connection. In this way, claimants can submit their applications and get compensations entirely online. On the other hand, the online as- sessment also helps the insurance company save expenses as the assessor can quickly approve quali- fied applications or screen out applications that do not conform to the claims on the online platform. However, limited by the complexity of the insur- ance …

Content truncated to 500 words