Design of Insurance Product Recommendation Model based on Bigdata and Artificial Intelligence

Abstract

Past research into insurance businesses is limited to current status. There is a lot to be desired about actively developing and sharing analytical models using Bigdata. The purpose of this study is to develop a model that recommends collateral and subscription amounts at the customer group level. First, we clustered our customers into a group of customers of similar economic size. We compared the collateral and subscription amount of the products customers in the cluster. This allowed the recommendation of collateral and subscription amounts to the level of customers in the same cluster. Customer mappings were mapped to clusters closest to the customer's cluster variables using the R Package 'Kohonen'. Through this research model with SOM-based Two Step Clustering and Collaborative Filtering, insurance companies can provide opportunities to internalize their analysis. In addition, it will be possible to improve the effectiveness of life insurance and ICT executives' work, and to support higher-level decisions using AI Tech.