FRAUD AND LATE DELIVERY PREDICTION USING HYBRID MODEL
The cargo sector is going through considerable expansion in volume owing to technical innovation in e-commerce and global trade liberalization. Volume expansion also indicates a surge in fraud cases involving smuggling and fraudulent reporting of goods. Shipping businesses and customs are largely dependent on normal random examination hence uncovering fraud is typically by coincidence. As the volume raises considerably it would no longer be viable and beneficial for both transportation firms and customs to pursue standard fraud detection tactics. Other related publications in this field have demonstrated that intelligent data-driven fraud detection is proved to be significantly more successful than regular inspections. The proposed system using machine learning algorithm for Support Vector Machine (SVM), Random Forest (RF) and Hybrid Scikit algorithms. As such in this article, we evaluate and then determine the most efficient methodologies and algorithms to detect fraud successfully within the shipping business. We also analyse characteristics that drive fraud activity, examine current fraud detection models, build the detection framework and apply the framework using the tool.
Fraud detection Models, Support Vector Machine (SVM), Random Forest (RF) and Hybrid Scikit algorithms.