MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641101083 A) filed by Dr. R. Aruna Kirithika on August 21, 2026, for "an Idsbd-Eai (intrusion Detection System In Big Data-Explainable Ai) Framework For Detection Of Intruders Based On Artificial Intelligence.
Inventors include Dr. R. Aruna Kirithika; Dr. S. Anand Christy; Mr. D. Vigneshwara Raj; Dr. J. Junia Deborah; Mrs. G. Uma; Dr. S. Muthukumaran; and Dr A. Victoria Anand Mary.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: Abstract The explosive growth of big data network environments has significantly expanded the attack surface for cyber threats, rendering traditional Intrusion Detection Systems (IDS) inadequate. While modern deep learning and Artificial Intelligence (AI) models offer unprecedented detection accuracy in parsing high- volume, high-velocity network traffic, their "black-box" nature creates a critical trust deficit. In high-stakes security operations, security analysts must not only know that an anomaly occurred, but why the AI flagged it. This research introduces IDSBD-EAI, a novel Intrusion Detection System framework optimized for Big Data environments, integrated with Explainable Al (XAI) capabilities. The framework aims to bridge the gap between high-performance anomaly detection and human- interpretable security forensics. Methodology: The proposed framework leverages a scalable big data processing pipeline (utilizing distributed frameworks like Apache Spark) to ingest and preprocess massive network traffic datasets in real time. The core detection engine employs an ensemble of advanced machine learning algorithms optimized for high-dimensional feature spaces. To demystify the model's predictions, an XAI layer utilizing SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) is integrated. This layer provides both global feature importance insights for system administrators and local, instance-level explanations for security analysts during active alerts. Results: Experimental validation conducted on benchmark big data security datasets (such as UNSW-NB15 or CICIDS2017) demonstrates that IDSBD-EAI achieves state-ofthe- art detection accuracy, precision, and low false-positive rates while maintaining high throughput. Crucially, the XAI layer successfully isolates the exact network attributes (e.g., packet length variance, protocol anomalies) driving each alert, reducing the mean time to detect and respond (MTTR) for security operations centres. Keywords: Intrusion Detection System (IDS); Big Data; Explainable Al (XAI); Cyber Security; Deep Learning Transparency; SHAP/LIME.
Disclaimer: Curated by HT Syndication.
