Machine Learning and ANN-Based Boat Booking Fraud Detection System

Abstract— Boat booking fraud endangers causes considerable loss for operators and honest customers alike. To manipulate booking and cancellation patterns for favoring illegal gains fraudsters employ sophisticated ways and exploit payment systems for the same. Our solution employs artificial neural networks and custom-built algorithms to analyze multidimensional booking data of transaction patterns, cancellation behavior, and lead time anomalies. It assesses customer signalling throughout the booking journey to identify abnormal patterns which are likely to escape human detection. By including domain-specific parameters such as seasonal booking trends, weather-connected cancellations and maritime payment behavior, our model achieves good accuracy in identifying fraudulent transactions during smooth operations with a low false positive rate. Not only protect service providers financially but also improve goodwill with customers through seamless solutions to genuine user booking processes. Implementation here shows how seamlessly those technologies which performed well in the hotel industry can be adapted to meet the exigencies of boat booking platforms. This is a part of a growing pool of specialized fraud detection systems in the maritime leisure market.