Fraud_Transaction_Detection

Fraudulent transactions using Machine Learning: Developed a model for predicting fraudulent transactions , which will predict fraud.Although by checking the count of Fraud and Legal transaction from the "is_Fraud" column, I got to know that the data is imbalanced .As count of Fraud Transaction is too less as compared to legal (no fraud ) transaction . So it is Imbalanced data ,as the No_Fraud Class has a very high number of observations and the Is_Fraud Class has a very low number of observations .Hence for an Imbalanced Class dataset F1 score is the most appropriate metric. Then the model which gives the best F1 score gives a more accurate result.