• Lalit Udayraj Pal
• Niraj Pramod Jain
• Santosh Singh
• Mithilesh Vishwakarma

The growing complexity of medical insurance claims has resulted in a notable increase in fraudulent activities, which in turn has led to substantial financial losses for insurance providers and has adversely affected healthcare costs overall. In light of this pressing issue, this study presents FraudNet, an innovative deep learning framework designed specifically to leverage cutting-edge machine learning techniques for the detection and mitigation of fraudulent claims within the realm of medical insurance. FraudNet utilizes a sophisticated combination of supervised and unsupervised learning methodologies, enabling it to effectively process a wide array of data sources. These sources include not only structured claim data but also unstructured textual information derived from claim narratives, which can often contain valuable insights. The model incorporates advanced algorithms such as K-Nearest Neighbors (KNN) and Support Vector Machines (SVMs) to uncover intricate patterns and anomalies that are typically associated with fraudulent behavior. The experimental results obtained from this research indicate that FraudNet significantly surpasses the performance of traditional fraud detection methods. It achieves a remarkably high accuracy rate while simultaneously reducing the occurrence of false positives, which is a common challenge in the field. This research makes a meaningful contribution to the expanding domain of artificial intelligence in healthcare, providing insurers with a powerful tool to enhance their fraud detection capabilities. Ultimately, this advancement holds the potential to foster a more sustainable healthcare system, benefiting both insurers and patients alike.

Medical Claim Insurance, Fraud Detection, Machine Learning, Data Preprocessing, Comparative Analysis

"Deep Learning Approaches for Detecting Fraudulent Claims in Medical Insurance", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 10, page no.e253-e264, October-2024, Available :http://www.jetir.org/papers/JETIR2410427.pdf

"Deep Learning Approaches for Detecting Fraudulent Claims in Medical Insurance", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 10, page no. ppe253-e264, October-2024, Available at : http://www.jetir.org/papers/JETIR2410427.pdf