Scholarlink

The fast growth of credit cards has sharply increased fraud cases. Conventional ways to detect fraud often do not pick up on time and transaction data patterns, when sensing affects other. Research on how LSTM networks may help uncover fraud. LSTMs are used for sequential data, e.g., language and music. Tests show that using our LSTM model 96.13% is a lot better than standard methods in use today. The model’s capacity to change to new scam patterns enhances its usefulness while reducing false positive and negative cases. The LSTM network result demonstrates the adaptability of the LSTM networks in allowing financial institutions to lessen losses and aid in conducting safe transactions.

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

Enhancing Fraud Detection and Transaction Security Through Optimized Pattern Recognition Using LSTM Networks

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • S. Sivakumar,
  • R. Jayanth,
  • U. Dharani,
  • M. Abishegan,
  • S. Nagul,
  • S. Harith Kumar Adhithya

摘要

The fast growth of credit cards has sharply increased fraud cases. Conventional ways to detect fraud often do not pick up on time and transaction data patterns, when sensing affects other. Research on how LSTM networks may help uncover fraud. LSTMs are used for sequential data, e.g., language and music. Tests show that using our LSTM model 96.13% is a lot better than standard methods in use today. The model’s capacity to change to new scam patterns enhances its usefulness while reducing false positive and negative cases. The LSTM network result demonstrates the adaptability of the LSTM networks in allowing financial institutions to lessen losses and aid in conducting safe transactions.