Anti Money Laundering detection using Naïve Bayes Classifier

Abstract

Anti Money laundering is a very challenging issue in the banking system. Anti-money laundering is defined as a set of procedures, policies and ordinances designed to prevent the creating income during illegal actions, e.g. market operations, deal of illegal commodities, and corruption of public funds and tax evasion. Our objective of the paper is to classify a transaction as illegal or not. To achieve this, we have used big data analytics technique for a dataset to identify the money laundering activities. We have used a customized dataset with 10000 transactions used for analysis with data cleaning, statistical analysis, and data mining process. The logical operator and if-else conditions are used to analyze the relationships among various attributes in the dataset prior to analyze with big data analytics methods. Finally, we have used the Naïve Bayes classifier to find money laundering activities. The analysis has been done using R software and customized dataset. The result obtained through analysis is very significant and proposed model achieved accuracy is 0.8125.