Fraud activities have been a serious threat to various sectors, including financial, insurance, and e-commerce. This wrong behavior leads to significant financial losses and damage to organizational reputation. This document presents a new and excellent model when detecting fraud when detecting fraud by using extended intellectual data analysis and machine learning methods to improve the accuracy and effect of the fraud detection system. This model includes preprocessing data, optimal function selection, and detection for detecting fraudulent behavior and patterns in large data. We explain the power of this model in the appendix for researchers in various global places, use real domain sets for academic seriousness, and to improve the use in the real world as the general indicators of the detection and decrease of false work. The final result is a reliable and expandable fraudulent detection system, which is against the attempt to deceive fraud with detection, financial crimes and inappropriate abuse of resources, to protect funds on behalf of other stakeholders, and to commit fraud in the region and sectors. In conclusion, this excellent model relies greatly on intellectual data analysis and machine learning, providing ideas for complex problems that detect fraud. In general, this model and other similar contributions will improve the organization of the organization in connection with the improvement of money or to alleviate the detected effects of financial interests in this way.