Financial Markets Fraud Detection and Prevention using Kernel Adatron Algorithm with Machine Learning

Abstract

The investors may have great interest to invest in stock market. Moreover, financial markets like stock markets are driven by explosive factors such as social media, micro blogs and news that make it hard to predict stock market index based on purely the historical data. The financial services industries that involve financial transactions are suffering from fraud- related losses and damages. Machine Learning (ML) is being rapidly adopted for a range of applications. It is important to begin considering the financial stability implications for every financial asset‟s organization. Using the machine learning tools and techniques in the finance sector will become necessary because it will closely monitor nascent and rapidly evolving landscape, wherein data on usage are largely unavailable, and bereft of any analysis. Financial assets fraud has seriously affected investors' confidence in the stock market and economic stability. The huge economic losses incurred because of several serious financial fraud events and because of this the intelligent financial fraud detection has thus been the topic of recent advances. In recent years, several studies have used stock market and machine learning techniques to provide solutions to this problem. In this paper, we propose various a state of art fraud detection techniques such as classification, clustering, and regression. This study aims to identify the techniques and methods that give the best results that have been perfected so far. Stock markets can benefit if fraud identification and prevention can be incorporated by using machine learning algorithms.