• Web traffic anomaly detection...

*Result*: Web traffic anomaly detection using deep learning techniques.

Title:
Web traffic anomaly detection using deep learning techniques.
Authors:
Anbcuchezhian, Ajay1 (AUTHOR) ajaya@karunya.edu.in, Paul, Malin Bruntha1 (AUTHOR) malin.bruntha@gmail.com, Rajeev, Rubin Ebenezer1 (AUTHOR) rubinebenezer@karunya.edu.in, Vallabadass, Bhuvanesh Immanuvel1 (AUTHOR) vbhuvanesh@karunya.edu.in, Raja, Vishwa1 (AUTHOR) vishwar20@karunya.edu.in, Sunadresan, Akila Agnes2 (AUTHOR) akila.s@gmrit.edu.in
Source:
AIP Conference Proceedings. 2025, Vol. 3291 Issue 1, p1-8. 8p.
Subject Terms:
*INTERNET traffic, *ANOMALY detection (Computer security), *TRAFFIC monitoring, *DEEP learning, *MALWARE
Database:
Academic Search Index

*Further Information*

*Web traffic refers to the data sent or received by users using online websites. Anomaly detection of Web traffic refers to sudden abnormal changes in traffic; this can lead to data manipulation, malware dissemination, and various other security threats. It is vital to perform anomaly detection accurately to avoid problems. This paper represents two models, RNN and LSTM, known for their ability to capture long-term dependencies and complex patterns from the raw data for real-time detection. The models were trained and tested on the S5 dataset from Yahoo. The Improved RNN architecture outperformed the LSTM model in anomaly detection, achieving an accuracy of 90.79% and recall of 87.4%. These results highlight the effectiveness of the Improved RNN model in web traffic anomaly detection, providing a valuable tool for enhancing online security. [ABSTRACT FROM AUTHOR]*