MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202621065126 A) filed by Sayali Prakash Shinde; Vishwakarma Institute Of Technology; Prof. Prajkta Dandavate; Aryan Gaikwad; Arnav Jain; Ashikaa Verma; and Vaidehi Pattarkine on May 23, 2026, for Hybrid Deep Neural Network And Ensemble Model For Predicting Consumer Preferences And Sales In E- Commerce Platforms.
Inventors include Prof. Prajkta Dandavate; Aryan Gaikwad; Arnav Jain; Ashikaa Verma; Vaidehi Pattarkine; and Prof Sayali Shinde.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: The rapid growth of e-commerce platforms has generated vast amounts of consumer behaviour data, creating opportunities to build intelligent systems for customer preference prediction and sales forecasting. Conventional machine learning methods struggle to capture complex, non-linear patterns in such data, while individual deep learning models often suffer from limited generalizability and robustness. To address these limitations, this paper proposes a Hybrid Deep Neural Network and Ensemble Model that combines a Deep Neural Network (DNN) with Random Forest (RF) and Gradient Boosting (GB) through a weighted aggregation strategy. The system is trained and evaluated on the publicly available UCI Online Shoppers Intention dataset (12,330 sessions, 18 features, 80%/20% train-test split). The proposed hybrid model achieves 95.3% accuracy, 96.1% precision, 95.7% recall, an F1-score of 95.9%, and an ROC-AUC of 0.983, outperforming all individual baseline models. Stress testing under 10–20% input noise confirmed robustness above 93% accuracy, with an average inference latency of under 120 ms. These results demonstrate that the proposed approach is well suited for real-time, large-scale e-commerce deployment.
Disclaimer: Curated by HT Syndication.
