Clustering of the population benefiting from health insurance using K-means
Sara Zahi, Boujemâa Achchab · Proceedings of the 4th International Conference on Smart City Applications · 2019
Clustering techniques aim to discover groupings of a set of patterns or data and are widely used in any discipline that involves analysis of multivariate data. Their applications in different fields are multiple and diverse. They can also be used in health insurance, which is an important part of healthcare. The use of clustering techniques in this field will provide opportunities and solutions for decision makers in order to monitor the insurance coverage in general and health insurance in particular. The aim of this paper is to create clusters of the insured population by means of an unsupervised machine learning algorithm, which is a partitional clustering technique that has proved to be efficient and fast, which is K-means. We present and discuss the results obtained by using this clustering method.