Framework for Explainable Feature Learning in Big Data Streams

Author(s)	Mr. Alex Dhoke, Mr. Nilesh Nagrale, Mr. Abhay Rewatkar
Country	India
Abstract	The proliferation of heterogeneous data streams from IoT, social media, and enterprise systems presents significant challenges for scalable and interpretable Big Data classification. While existing approaches often prioritize predictive accuracy over transparency, post-hoc explanation methods such as LIME and SHAP are computationally prohibitive in streaming contexts. Recent studies advocate for intrinsic interpretability via attention mechanisms and disentangled representations, yet their application to multimodal, high-velocity streams remains underexplored. This paper proposes an integrated framework for explainable feature learning that embeds interpretability directly into the classification pipeline. The framework leverages multimodal autoencoders and attention mechanisms to generate semantically meaningful features and real-time explanations. Evaluations on multimodal streaming datasets demonstrate competitive predictive performance and significant improvements in explanation fidelity, making the approach suitable for regulated domains such as healthcare and cybersecurity.
Keywords	Explainable AI, Big Data, Heterogeneous Data Streams, Autoencoders, Attention Mechanisms, Real-Time Classification.
Published In	Volume 7, Issue 2, March-April 2026
Published On	2026-04-24
DOI	https://doi.org/10.63363/aijfr.2026.v07i02.5186

About AIJFR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9898-79-39-59	editor@aijfr.com

Advanced International Journal for Research