• A.KALAIVANI
• R. Pugazendi

Intrusion Detection Systems plays an vital role for protecting the computer networks from the trending cyberattacks which is increasing drastically .The traditional machine learning based IDS is lacking and has many drawbacks in many ways. The machine learning based IDS completely depends heavily on the handcrafted features; it needs extensive preprocessing and requires explicit normalization to be done. This limits the ML models for IDS their ability to adapt to evolving attack patterns. The limitations this work focuses based on a deep learning-based intrusion detection system. This end to end deep learning based IDS operates directly on raw flow-level network traffic with minimal preprocessing. A Convolutional Neural Network (CNN) is employed to automatically learn the discriminative feature representations and perform attack classification. This paper proposes a model by selecting raw traffic selection included in the CSE-CIC-IDS2018 dataset and batch normalization layers are embedded within the CNN which handles the internal feature scaling and it also eliminates the need for the explicit data normalization. The proposed framework achieved an accuracy of 95.6% which has outperformed the traditional machine learning models. Thus this result establishes a strong baseline for IDS and it also motivates further using advanced feature learning and optimization strategies.

Intrusion Detection System, Deep Learning, CNN, Automatic Feature

"End-to-End Deep Learning Framework for Intrusion Detection Using Raw Network Traffic Data", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 2, page no.b692-b699, February-2026, Available :http://www.jetir.org/papers/JETIR2602190.pdf

"End-to-End Deep Learning Framework for Intrusion Detection Using Raw Network Traffic Data", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 2, page no. ppb692-b699, February-2026, Available at : http://www.jetir.org/papers/JETIR2602190.pdf