REVOLUTIONARY INTRUSION DETECTION: HFS-MLSTM WITH HYBRID FEATURE SELECTION AND ATTENTION-DRIVEN BILSTM MODEL

REVOLUTIONARY INTRUSION DETECTION: HFS-MLSTM WITH HYBRID FEATURE SELECTION AND ATTENTION-DRIVEN BILSTM MODEL J.Archana

Intrusion detection systems (IDS) are critical for maintaining the security of network environments. The increasing sophistication and frequency of cyber-attacks necessitate the development of advanced IDS capable of identifying and mitigating threats in real-time. Traditional IDS models often struggle with high false positive rates, computational inefficiencies, and the inability to adapt to evolving attack patterns. These limitations highlight the need for more robust and intelligent detection mechanisms. This paper introduces a novel approach named HFS-MLSTM (Hybrid Feature Selection and Multi-Layer Long Short-Term Memory) integrated with an attention-driven BiLSTM (Bidirectional Long Short-Term Memory) model to enhance intrusion detection (ID) capabilities. The proposed method leverages hybrid feature selection techniques, combining Mutual Information (MI) and Recursive Feature Elimination (RFE), to identify the most relevant features for accurate detection. The BiLSTM model, augmented with an attention mechanism, improves the model’s ability to focus on critical aspects of the input data, leading to more precise detection of anomalous activities. We have evaluated the effectiveness of our HFS-MLSTM using standard benchmark datasets and shown that it has outperformed traditional IDS approaches by 99% accuracy. This innovative model offers a promising solution for real-time ID, ensuring robust network security and resilience against evolving cyber threats.
10 22 2025 146 179 10.12732/ijam.v38i7s.472 https://ijamjournal.org/ijam/publication/index.php/ijam/article/view/472 https://ijamjournal.org/ijam/publication/index.php/ijam/article/download/472/433 https://ijamjournal.org/ijam/publication/index.php/ijam/article/download/472/433