Thesis Abstract

Abstract
This thesis presents the development of a predictive model for fraud detection in insurance claims. The insurance industry faces significant challenges in detecting and preventing fraudulent activities, which can result in substantial financial losses. Traditional methods of fraud detection often fall short in accurately identifying fraudulent claims, leading to increased costs for insurers and higher premiums for policyholders. To address this issue, this research focuses on the implementation of a predictive model that leverages advanced machine learning algorithms to effectively detect fraudulent insurance claims. The study begins with a comprehensive review of existing literature on fraud detection in the insurance industry. This review highlights the limitations of current methods and emphasizes the need for more sophisticated techniques to combat fraud effectively. The research methodology section outlines the data collection process, feature selection techniques, and model development process. Various machine learning algorithms, including logistic regression, decision trees, random forest, and neural networks, are explored and compared to identify the most effective approach for fraud detection in insurance claims. The findings of the study reveal that the predictive model developed using ensemble learning techniques, specifically the random forest algorithm, outperforms other models in terms of accuracy and efficiency. The model demonstrates a high level of sensitivity and specificity in detecting fraudulent claims, significantly reducing false positives and false negatives. The discussion of findings section analyzes the performance metrics of the model and provides insights into the factors influencing fraud detection accuracy. In conclusion, the development of a predictive model for fraud detection in insurance claims represents a significant advancement in the field of insurance fraud prevention. By leveraging advanced machine learning techniques, insurers can enhance their ability to identify and prevent fraudulent activities, ultimately leading to cost savings and improved customer satisfaction. The study contributes to the body of knowledge on fraud detection in the insurance industry and provides practical implications for insurers looking to enhance their fraud detection capabilities. Keywords Fraud detection, Insurance claims, Predictive modeling, Machine learning, Random forest, Ensemble learning

Thesis Overview