Authors: Shaghayegh Abolmakarem; Farshid Abdi; Kaveh Khalili-Damghani
Addresses: Department of Industrial Engineering, South-Tehran Branch, Islamic Azad University, Tehran, Iran ' Department of Industrial Engineering, South-Tehran Branch, Islamic Azad University, Tehran, Iran ' Department of Industrial Engineering, South-Tehran Branch, Islamic Azad University, Tehran, Iran
Abstract: Customers segmentation enables companies to identify the high-profit customers. Clustering algorithms are commonly used for customer segmentation. In this study, K-means clustering algorithms are employed to identify profitable customers in an insurance company. The optimum number of clusters is determined using 'NbClust' package in R software through calculating 23 clustering evaluation metrics. The clustering is accomplished on insurance customers on the basis of 16 customers' features, and ten insurance feature using CRISP methodology. The results show that the customers of insurance company are divided into three groups labelled as 'profitable customers', 'potential profitable customers', and 'disinterested customers'. On the basis of the results of this study, associated customer relationship management (CRM) strategies are proposed to establish suitable marketing and communication plans for each cluster of customers.
Keywords: data mining; k-means clustering; validity indices; customer grouping; profitable customers; insurance customers; customer segmentation; customer relationship management; CRM strategies; marketing plans; communication plans.
International Journal of Knowledge Engineering and Data Mining, 2016 Vol.4 No.1, pp.18 - 39
Available online: 06 Feb 2017 *Full-text access for editors Access for subscribers Purchase this article Comment on this article