Solving customer insurance coverage recommendation problem using a two-stage clustering-classification model

Abstract

Insurance coverage recommendation problem (ICRP) in which the most suitable coverage for customers is suggested is an essential issue for an insurance company. ICRP helps insurance companies to give suitable services to their customers. In ICRP, the insurance company tried to mine the features and records of data associated with the customers to suggest them the most economic and fitted insurance plan. The insurance companies have large databases which are considered as a proper infrastructure to analyze, model and predict the customer behavior. In this paper, a two-stage clustering-classification model is proposed to suggest suitable insurance coverage for customers. The first stage addresses a data pre-screening phase and clustering of customers based on the record of insurance coverage. Well-known clustering algorithms are employed. The superior clustering algorithm is selected based on Davies-Bouldin metric. In the second stage, several filter and wrapper methods are implemented to select proper features. The selected features are assumed as inputs of K-nearest neighbor classification algorithm. The proposed approach is applied in a real case study for clustering the customers and recommend insurance coverage. The results show that the model is capable of suggesting suitable insurance coverage based on customers’ characteristics.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.