UPI Fraud Transaction Detection using Machine Learning || VIP || IEEE || HYD || APPS || PROJECTS

TO PURCHASE OUR PROJECTS IN ONLINE (OR) OFFLINE
CONTACT:VENKAT INNOVATIVE PROJECTS
NAME: VENKATARAO GANIPISETTY
Mobile & WhatsApp :[phone]
Mobile & WhatsApp :[phone]
Email :[e-mail]
Email :[e-mail]
Website:venkatprojects.com
website:venkatinnovati...
ABOUT PROJECT:
Growing technologies converting all manual works to virtual works such as online shopping or online transaction where users can shop online and can make payments online. This advantage leads to a problem called Fraud transactions where malicious attackers can perform phishing activities to generate fraud transaction. All banks investing billions of money to detect and prevent such frauds attacks but no algorithms are giving accurate prediction.
In propose work we are employing advance machine learning algorithm such as XGBOOST which has a group of estimators and decision trees for accurate prediction. This algorithm trained on fraud transaction dataset and manages to detect fraud transaction on test data with an accuracy of over 99%.
To enhance algorithm performance we have employed Principal Component Analysis (PCA) algorithm to select relevant features from the dataset and then selected features will be input to XGBOOST to train a model.
For training we are using 70% dataset and for testing we have utilize 30% dataset. We have experimented with existing Naïve Bayes algorithm and propose XGBOOST algorithm. Each algorithm performance is measured by using different metrics such as Confusion Matrix, Accuracy, Precision, Recall and FSCORE.