Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{175226,
        author = {SHAISTA SULTANA and Dr M Swapna},
        title = {DeepFusion: A Robust Framework for Deepfake Video Classification Using Convolutional and Recurrent Neural Networks},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {11},
        pages = {2148-2155},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=175226},
        abstract = {Deepfake videos, developed using sophisticated artificial intelligence methods, challenge the credibility and security of digital content by fabricating highly realistic but deceptive visuals. To counter this, a hybrid deep learning framework has been implemented, combining InceptionV3 for capturing detailed spatial features from video frames with Gated Recurrent Units (GRUs) for recognizing sequential temporal patterns. The approach involves preprocessing video data, extracting intricate frame-level features, and analyzing temporal consistencies through GRUs to identify manipulations effectively. Binary cross- entropy loss directs the training process, with early stopping mechanisms ensuring robust and efficient learning. Tested on a curated dataset, this method provides a reliable solution for preserving the authenticity of multimedia content and addressing the risks posed by deepfake technologies.},
        keywords = {Deepfake Detection, InceptionV3, GRU, Temporal Dependencies, Feature Extraction, Video Classification, Deep Learning, Multimedia Content Integrity, AI Algorithms, Convolutional Neural Networks, Recurrent Neural Networks, Early Stopping, Binary Cross-Entropy, Digital Security, Media Forensics.},
        month = {April},
        }