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{204256,
        author = {Mrs.M.Nandhiya and Dr.C.R Durga devi},
        title = {IoT Enabled Robust Deep Capsule Network for Cloud-Based Predictive Analytics},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {1},
        pages = {4099-4102},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=204256},
        abstract = {Cloud computing provides on-demand access to scalable resources, including storage, networking, analytics, and software applications through internet-based platforms. Predictive analytics has become a significant cloud computing application, enabling the estimation of future outcomes using historical and real-time data. However, conventional prediction techniques often experience reduced accuracy and increased computational overhead when processing large-scale and heterogeneous datasets.
This study proposes an IoT-enabled Robust Autoencoded Deep Capsule Network to enhance predictive analytics in cloud computing environments. The developed framework combines data preprocessing, feature extraction, classification, and optimization within an integrated architecture. IoT devices continuously collect environmental information and transmit it to cloud servers for further processing and analysis. During the preprocessing stage, missing and anomalous data are addressed using polynomial-based imputation and modified Z-score techniques. Subsequently, a robust autoencoder extracts significant features, thereby reducing data dimensionality and enhancing computational efficiency. The extracted features are then processed by a capsule network that employs a similarity-driven classification mechanism to produce reliable prediction outcomes.
Performance assessment conducted on the Global Air Pollution Dataset indicates that the proposed approach outperforms existing deep learning models in terms of prediction accuracy while simultaneously reducing prediction errors and execution time.},
        keywords = {IoT, Cloud Computing, Predictive Analytics, Deep Capsule Network, Autoencoder, Feature Selection, Air Pollution Prediction, Machine Learning},
        month = {June},
        }