Customs Fraud Detection Method Based on Artificial Intelligence and Data Analytics

Because of the increased market opening and massive data collection brought about by globalization, the need of maintaining control over customs procedures has increased. However, the integration and processing of customs data are complicated by its extreme imbalance. Finding autonomous, computationally sophisticated solutions for customs handling is therefore crucial. The study described in this paper aims to propose an artificial intelligence-based method for customs fraud detection through data analytics. A machine learning-based approach for detecting customs fraud is proposed. The research methodology includes data preprocessing, data analytics, combining results, and deriving precise solutions from results obtained. The aim is to extract meaningful information from a dataset. For the purpose of classifying customs fraud, seven algorithms for machine-learning-kNN, SGD, Constant, Logistic Regression, Random Forest, Neural Network, and Naive Bayes-are chosen and empirically assessed.