Sentiment Analysis of Long-Term Care Insurance in China Using NLP

Yixuan Li

doi:10.62051/dp3cev65

Authors

Yixuan Li

DOI:

https://doi.org/10.62051/dp3cev65

Keywords:

Long-Term Care Insurance (LTCI); Natural Language Processing (NLP); Sentiment Trends; Deep Learning.

Abstract

Long-term care insurance (LTCI) is important for addressing the needs of aging populations and increasing long-term care demand. China initiated its LTCI pilot in 2016 and expanded it in 2019. This study employs Natural Language Processing (NLP) to analyze the sentiment impact of policyholders on LTCI decisions. By classifying sentiments in comments from various Chinese provinces, this study trained a Long Short-Term Memory (LSTM)-based deep learning model using annotated data and validated its accuracy with a pre-extracted test dataset. The application of the trained model to LTCI comments revealed sentiment distributions across different provinces, highlighting the developmental progress of LTCI since its pilot implementation. This research not only demonstrates the effective use of NLP in understanding sentiment trends but also underscores the significance of transitioning from qualitative to quantitative analysis. The study proposes integrating NLP with behavioral finance and insurance disciplines to enhance the utility and value of insurance products. This approach can guide product development, refine target marketing strategies, and inform policy decisions, contributing to the future expansion and success of LTCI in China.

Downloads

Download data is not yet available.

References

[1] Lerner J.S. Small D.A. Loewenstein G. Heart strings and purse strings: Carryover effects of emotions on economic decisions. Psychological science, 2004, 15 (5): 337 - 341.

[2] Kahneman D. Tversky A. Prospect theory: An analysis of decision under risk. Handbook of the fundamentals of financial decision making, 2013: 99 - 127.

[3] Kahneman D. Sugden R. Experienced utility as a standard of policy evaluation. Environmental and resource economics, 2005, 32: 161 - 181.

[4] Bracha A. Brown D.J. Affective decision making: A theory of optimism bias. Games and Economic Behavior, 2012, 75 (1): 67 - 80.

[5] Slovic P. Finucane M.L. Peters E. et al. Risk as analysis and risk as feelings: Some thoughts about affect, reason, risk and rationality. The feeling of risk. Routledge, 2013: 21 - 36.

[6] Tillmann C. Vogel S. Ney H. et al. Accelerated DP based search for statistical translation. Eurospeech, 1997: 2667 - 2670.

[7] McCallum A. Nigam K. A comparison of event models for naive bayes text classification. In AAAI-98 workshop on learning for text categorization, 1998, 752: 41 - 48.

[8] Hochreiter S. Schmidhuber J. Long short-term memory. Neural Comput, 1997, 9 (8): 1735 – 1780.

[9] Devlin J. Chang M.W. Lee K. Toutanova K. Bert: pre-training of deep bidirectional transformers for language understanding. 2018, arXiv preprint: 1810.04805.

[10] Rospocher M. van Erp M. Vossen P. Fokkens A. Aldabe I. Rigau G. Soroa A. Ploeger T. Bogaard T. Building event-centric knowledge graphs from news. Web Semantics: Science, Services and Agents on the World Wide Web, 2016.

[11] Sharma S. Srinivas P.Y.K.L. Balabantaray R.C. Emotion Detection using Online Machine Learning Method and TLBO on Mixed Script. In Proceedings of Language Resources and Evaluation Conference, 2016: 47 - 51.

[12] Seal D. Roy U.K. Basak R. Sentence-level emotion detection from text based on semantic rules. Advances in intelligent Systems and computing, 2020, 933.

[13] Kaggle dataset, “Weibo comment data”, 2023, https://www.kaggle.com/datasets/yisywanlee/weibo-comments.

[14] Kaggle dataset, “final data”, 2023, https://www.kaggle.com/datasets/yisywanlee/weibo-comments-regarding-long-term-care-insurance.