Machine Learning Statisitical Analysis Fraud Detection Data Science Statisitcs
Abstract:
Machine learning is one of the newest and most effective ways to classify data and make predictions. There are numerous machine learning methods that are used for a variety of tasks, such as prediction and classification. Before machine learning, statistical analysis was the main method used to make predictions and classify data. Like machine learning, statistical analysis has an abundance of procedures that can be used. This thesis aims to compare the old and the new: machine learning with statistical procedures. The goal of this study is to detect fraudulent transactions amongst a financial data set. Machine learning and statistical analysis each offer a unique method of addressing the situation. Through a review of the literature, experimentation and analysis, this study aims to thoroughly examine various approaches, with a particular emphasis on how well they identify fraudulent activity in financial transactions. The results of this thesis provide important new information about how machine learning and statistical methods compare in terms of their ability to detect fraud in financial transactions. This research intends to enhance decision-making processes in the implementation of fraud detection systems and inspire further breakthroughs in machine learning and statistical approaches for efficiently combating financial crime by illuminating the advantages and disadvantages of each strategy.
