ChaoG-GRU: Healthcare Transaction Fraud Detection using Chaotic Iterative Gazelle-based Gated Recurrent Unit Technique

This research presents an innovative approach to fraud detection in healthcare transactions by combining imputation techniques and a novel algorithm. The study begins by addressing the challenge of missing data in the dataset through the application of the K-Nearest Neighbour (KNN) imputation method. Subsequently, the research employs a specialized fraud detection mechanism utilizing the Chaotic Iterative Gazelle-based Gated Recurrent Unit (ChaoG-GRU). The ChaoG is a unique construct that integrates chaotic iterative mapping into the Gazelle optimization, specifically designed to address issues related to local optimal trapping. The introduction of the ChaoG-GRU algorithm represents a significant advancement in fraud detection techniques. The integration of Chaotic Iterative Gazelle into the Gated Recurrent Unit architecture enhances the model's ability to capture intricate patterns within healthcare transactions, providing a more sophisticated and effective approach to fraud detection.