資料載入中.....
|
請使用永久網址來引用或連結此文件:
https://nccur.lib.nccu.edu.tw/handle/140.119/152768
|
| 題名: | 智慧保險理賠評估暨推薦系統設計以醫療險為例
Design of an Intelligent Insurance Claim Assessment and Recommendation System: Case Study on Health Insurance |
| 作者: | 翁世育
Weng, Shih-Yu |
| 貢獻者: | 蔡子傑
Tsai, Tzu-Chieh
翁世育
Weng, Shih-Yu |
| 關鍵詞: | 人工智慧
大型語言模型
檢索增強生成
保險賠償
語意辨識
資訊安全
Artificial Intelligence
Large language model
RAG
Insurance compensation
Semantic recognition
Information security |
| 日期: | 2024 |
| 上傳時間: | 2024-08-05 13:55:59 (UTC+8) |
| 摘要: | 鑑於商業保險市場迅速發展，許多人擁有多份保單，這些保單隨著時間更新，條款也不斷調整。由於保險條款的專業性，保戶在理解上常需依賴保險專家或業務員的建議。然而，業務員也可能無法記住所有條款，需要諮詢公司專業人員，這會耗費時間且缺乏即時性，或因利益衝突或不當得利，導致理賠糾紛。為減少糾紛，保戶需詳細了解保單條款、保障範圍、除外條款及賠償限制等內容，並尋求專業建議。
為解決這些問題，運用大語言模型和RAG技術，構建交談式保險理賠暨推薦系統。該系統能解析保單條款，分析保戶的理賠問題，通過關鍵字提取並計算條款匹配的信心分數，將結果回傳並以易懂的方式回答保戶，協助他們輕鬆獲得相應答覆。系統將複雜保險條款轉換為易於理解的語言，降低理賠過程的困難和不確定性。
系統設計需考慮許多因素，包括數據正確性、演算法效率、友善地使用者介面及隱私安全問題。系統需持續更新，以適應保險市場變化、新法規、新產品和技術標準等，最終，這種交談式保險
理賠暨推薦系統能提高理賠效率和準確性，減少理賠糾紛，保護保戶權益。同時，也能幫助保險公司提升客戶服務質量和滿意度，提高業務效率並降低成本，為保戶和保險公司帶來多重益處。
Due to the rapid development of the commercial insurance market, many people hold multiple policies that are updated over time, with terms that are constantly adjusted. Because of the specialized nature of insurance terms, policyholders often need to rely on the advice of insurance experts or agents for understanding. However, agents may not remember all the terms and might need to consult company specialists, which can be time-consuming and lack immediacy. Additionally, conflicts of interest or improper gains can lead to claims disputes. To reduce such disputes, policyholders need to thoroughly understand the policy terms, coverage scope, exclusions, and compensation limits, and seek professional advice.
To address these issues, a conversational insurance claims and recommendation system using large language models and Retrieval-Augmented Generation (RAG) technology can be developed. This system can interpret policy terms, analyze policyholder claims issues, extract keywords, and calculate the confidence scores of term matches. The results are then returned and presented in an understandable manner, assisting policyholders in obtaining relevant answers easily. The system translates complex insurance terms into easy-to-understand language, reducing the difficulty and uncertainty of the claims process.
The system design needs to consider many factors, including data accuracy, algorithm efficiency, user interface friendliness, and privacy and security issues. The system must be continuously updated to adapt to changes in the insurance market, new regulations, new products, and technical standards. Ultimately, this conversational insurance claims and recommendation system can improve claims efficiency and accuracy, reduce claims disputes, and protect policyholders' rights. At the same time, it can help insurance companies enhance customer service quality and satisfaction, improve business efficiency, and reduce costs, bringing multiple benefits to both policyholders and insurance companies. |
| 參考文獻: | 1. Jacob Devlin, Ming-Wei Chang, Kenton Lee, Kristina Toutanova (2019) BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding https://arxiv.org/abs/1810.04805
2. Bin Fu, Yunqi Qiu, Chengguang Tang,(2020) A Survey on Complex Question Answering over Knowledge Base: Recent Advances and Challenges https://arxiv.org/abs/2007.13069
3. Seon-Ok Na, Young-Min Kim, Seung-Hwan Cho (2022) Insurance Question Answering via Single-turn Dialogue Modeling https://aclanthology.org/2022.cai-1.5/
4. Wayne Xin Zhao, Kun Zhou,(2023)A Survey of Large Language Models https://arxiv.org/pdf/2303.18223
5. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert,(2023) Llama 2: Open Foundation and Fine-Tuned Chat Models https://arxiv.org/abs/2307.09288
6. Gemma Team: Thomas Mesnard, Cassidy Hardin, Robert Dadashi,(2024) Gemma: Open Models Based on Gemini Research and Technology https://arxiv.org/abs/2403.08295
7. Marah Abdin, Sam Ade Jacobs, Ammar Ahmad Awan,(2024) Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone https://arxiv.org/abs/2404.14219
8. Patrick Lewis, Ethan Perez, Aleksandra Piktus,(2020), Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks https://arxiv.org/abs/2005.11401
9. Yunfan Gao, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yi Dai, Jiawei Sun, Meng Wang, Haofen Wang, (2023) Retrieval-Augmented Generation for Large Language Models: A Survey https://arxiv.org/abs/2312.10997
10. Bertalan Mesko , (2023) The imperative for regulatory oversight of large language models (or generative AI) in healthcare https://www.nature.com/articles/s41746-023-00873-0
11. Manuel Lentzen; Thomas Linden; Sai Veeranki; Sumit Madan (2023) A Transformer-Based Model Trained on Large Scale Claims Data for PredictionofSevere COVID-19 Disease Progression https://ieeexplore.ieee.org/document/10159467
12. Frank Kaptein, Bernd Kiefer, Antoine Cully, (2021) A Cloud-based Robot System for Long-term Interaction: Principles, Implementation, Lessons Learned https://dl.acm.org/doi/10.1145/3481585
13. Yiming Cui, Ziqing Yang, Xin Yao (2023) Efficient and Effective Text Encoding for Chinese LLaMA and Alpaca https://arxiv.org/abs/2304.08177
14. Xinyu Lin, Wenjie Wang, Yongqi Li,(2024) Data-efficient Fine-tuning for LLM-based Recommendation https://arxiv.org/abs/2401.17197
15. Ranjodh Singh; Meghna P. Ayyar; Tata Venkata Sri Pavan; (2019) AUTOMATING CAR INSURANCE CLAIMS USING DEEP LEARNING TECHNIQUES https://ieeexplore.ieee.org/document/8919258
16. Gareth William Peters, Efstathios Panayi (2015). Understanding Modern Banking Ledgers through Blockchain Technologies: Future of Transaction Processing and Smart Contracts on the Internet of Money https://arxiv.org/abs/1511.05740
17. Lin, Chan Min(2016), Streamlining Medical Insurance Claims Processing With Smart Contracts 政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/100573
18. Banghao Chen, Zhaofeng Zhang, Nicolas Langrené, (2024) Unleashing the potential of prompt engineering: a comprehensive review https://arxiv.org/abs/2310.14735
19. Xavier Amatriain (2024)Prompt Engineering a Prompt Engineer https://arxiv.org/abs/2401.14423
20. 財團法人保險事業發展中心 (tii.org.tw)
21. 全民健康保險醫療服務給付項目及支付標準
22. 財團法人金融消費評議中心
23. 衛生福利部全民健康保險爭議審議會
24. Lemonade: An Insurance Company Built for the 21st Century
25. MY83 保險網－把保險變簡單
26. 智能保險規劃 – 免費獲得專屬規劃｜全球人壽 (transglobe.com.tw) |
| 描述: | 碩士
國立政治大學
資訊科學系碩士在職專班
110971030 |
| 資料來源: | http://thesis.lib.nccu.edu.tw/record/#G0110971030 |
| 資料類型: | thesis |
| 顯示於類別: | [資訊科學系碩士在職專班] 學位論文
|
文件中的檔案:
| 檔案 |
描述 |
大小 | 格式 | 瀏覽次數 |
|---|
| 103001.pdf | | 3598Kb | Adobe PDF | 0 | 檢視/開啟 |
|
在政大典藏中所有的資料項目都受到原著作權保護.
|
