Towards Better Detection of Fraud in Health Insurance Claims in Kenya: Use of Naïve Bayes Classification Algorithm

The extent, possibility, and complexity of the healthcare industry have attracted widespread fraud that has contributed to rising costs hence affecting patients’ health negatively impacting economy many countries. Despite putting up various technologies strategies fight such as planned, targeted audits, random whistle-blowing, biometric systems, in claims continued be a challenge most insurance providers Kenya. This paper explored application data mining detecting Classification algorithms (Naïve Bayes, Decision Tree K-Nearest Neighbour) were used build predictive models for knowledge discovery process. After conducting several experiments, resulting showed Naïve Bayes works well with 91.790% classification accuracy 74.12% testing hit rate. A prototype was developed based on rules extracted from model, which, if adopted, will save by it is committed. Fraud detection much needed countries so help reduce loss money return improve service delivery patients.