Fraud Analytics in the Insurance Sector

Fraud Analytics in the Insurance Sector

Author Name : Gayatri Bharat Magar, Jay Manohar Walkunde, Saurabh Ramesh Waghmare, Vinit Bharat Gowardhan, Prof. Jitendra Musale, Prof. Shweta Joshi

ABSTRACT Over the past few years, an insurance company that operates as a business has seen fraud instances involving various kinds of claims. Since the amount claimed by fraudsters is so large and could lead to major issues, various organizations are working with the government to identify and curtail these kinds of operations. Nowadays, fraud is a major problem facing the insurance industry that Big Data and Machine Learning are trying to solve. This paper deals with the evaluation of the effectiveness and the verifiability of the best-known machine learning algorithms for fraud prediction. We adopted the supervised method applied to automobile data claims of an anonymous insurance company. We aim to propose an approach that improves the relevance of the results of artificial intelligence. Additionally, to compare all machine learning methods that are utilized for classification utilizing confusion matrices in terms of recall, soft accuracy, and precision, among other metrics. The study has demonstrated that Logistic Regression works better among all algorithms compared.As the different countries around the world evolve into a more economical-based and stimulating their economy is the goal. The main purpose of most of these countries is to fight off money launderers and fraudsters for better economic growth. A popular fraud topic in this regard is insurance fraud since it costs the companies and the public billions. Applying data analysis and machine learning are great ways used to address many problems regarding any automated system. them out in saving money and time and helping them become more efficient in reacting to these fraudulentclaims.