Predictive analysis and supervised detection for fraudulent cases in healthcare

Abstract

Healthcare is perhaps one of the most crucial industry for humanity and it has been, and it will always be a growing industry. This increases the risk of being exploited. In our paper, we performed and analyzed different trends for the suspicious/fraudulent medical activities. We categorized different groups of patients involved and analyzed their distribution on the basis of multiple factors. Healthcare is a massive and widely distributed sector with a numerous entities and stakeholders involved. Limited connectivity within these distributed management create various loopholes that people try to exploit. We performed multiple analysis for suspicious and fraudulent activities in the healthcare/Medicare industry and tried to look for popular trends people opt to exploit the system. We also tried various supervised machine learning algorithms to see how they behave with our dataset and what is the accuracy of their detection. This can be helpful in choosing the best model while building a solution to deal with various use cases involved in fraudulent activity detection. Healthcare is one of the sectors that will always be relevant to living beings and should be taken special care of.