Scholarlink

In today’s world, online payment is mostly used, making it easy for fraudsters to cause fraud in banking. Mobile payment scams are a significant concern in the unbounded world economy, affecting millions of transactions. This study employs the PaySim dataset, comprising over one million simulated mobile financial records, to develop a robust fraud detection system using ML techniques. A RF classifier outperforms the contenders like SVM, Logistic Regression, and Naïve Bayes as it scores 97.59% accuracy, 96.30% precision, 97% recall, and an F1-score of 98%. This scalable solution demonstrates that with the help of machine learning, highly accurate fraudulent transactions can be prevented in factors for decision criteria in mobile payments.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Scalable Fraud Detection in Mobile Payment Transactions Using Leveraging Machine-Learning Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Swapnil Patil,
  • Pradyumna Shukla,
  • Mahak Shah,
  • Akaash Vishal Hazarika

摘要

In today’s world, online payment is mostly used, making it easy for fraudsters to cause fraud in banking. Mobile payment scams are a significant concern in the unbounded world economy, affecting millions of transactions. This study employs the PaySim dataset, comprising over one million simulated mobile financial records, to develop a robust fraud detection system using ML techniques. A RF classifier outperforms the contenders like SVM, Logistic Regression, and Naïve Bayes as it scores 97.59% accuracy, 96.30% precision, 97% recall, and an F1-score of 98%. This scalable solution demonstrates that with the help of machine learning, highly accurate fraudulent transactions can be prevented in factors for decision criteria in mobile payments.