Enhancing Enterprise Financial Fraud Detection Using Machine Learning

Abstract

The aim of their research is to improve the detection of financial fraud in enterprises through the utilization of artificial intelligence (AI) methodologies. The framework employs machine learning algorithms and data analytics to accurately identify patterns, anomalies, and signs of fraudulent activity. They employed exploratory data analysis approaches to identify instances of missing values and imbalanced data. The selection of the Random Forest Classifier is based on its ability to consistently capture intricate patterns and efficiently tackle the problem of multicollinearity. The isolation forest approach yielded an accuracy of 99.7%, while the local outlier factor method achieved an accuracy of 99.8%. Similarly, the Random Forest algorithm demonstrated an accuracy of 99.9%. The objective of their study is to aid organizations in proactively identifying instances of fraud by utilizing artificial intelligence methodologies.