Income-Tax-Fraud-Detection

Our Objective is to use backtracking anomaly detection machine learning concepts to detect the possible income tax fraud This will first eliminate the reductant transactions and then identity any possible sharp practice involved in the transactions. Our Idea also include identification of middleman /brokers involved in the fraud transaction and a severity index will be provided by our app. Firts anamoly detection is performed using one class SVM which is then passed to node js backend where using backtraing the fraud is confirmed.The anomaly detected along with their related transactions are stored in the blockchain to make this process immutable Machine learning is Implemented in Python integrated in Django Node Js is used to implement Blockchain