MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641087213 A) filed by Sr University, Warangal on July 16, 2026, for A System And Method For Adaptive Hybrid Ensemble Malware Detection Using Dynamic Feature Fusion And Machine–deep Learning Classifiers.
Inventors include Mrs. M. Sravani; and Dr. R. Vijaya Prakash.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: The present system discloses a system and method for adaptive hybrid ensemble malware detection using dynamic feature fusion and machine–deep learning classifiers for accurate, scalable, and real-time identification of malicious software across heterogeneous computing environments. The invention comprises a malware acquisition module configured to collect static, dynamic, behavioral, network, and contextual data from executable files, system logs, memory traces, and network traffic. A Dynamic Feature Fusion (DFF) engine preprocesses, normalizes, and adaptively combines heterogeneous feature sets using relevance-based feature weighting and contextual feature selection to generate an optimized feature representation. The optimized feature vector is processed in parallel by a plurality of machine learning classifiers, including Random Forest, Support Vector Machine, XGBoost, and LightGBM, and deep learning classifiers, including Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM), Gated Recurrent Unit (GRU), Transformer Encoder, and Autoencoder models. An Adaptive Hybrid Ensemble Learning (AHEL) module dynamically assigns confidence-based weights to individual classifiers according to prediction confidence, historical performance, and robustness, and integrates the classifier outputs through an Adaptive Confidence-Based Decision Fusion (ACDF) mechanism to generate a final malware classification. An Explainable Artificial Intelligence (XAI) module provides interpretable feature importance and classifier contribution analysis for transparent decision-making. The proposed invention improves detection accuracy for known, unknown, polymorphic, metamorphic, and zero-day malware while reducing false positives, computational overhead, and response latency. The framework is suitable for deployment in cloud computing, edge computing, enterprise networks, industrial control systems, healthcare systems, and Internet of Things (IoT) environments.
Disclaimer: Curated by HT Syndication.
