Intellectual Property India Publishes Patent Application for 'A Method And Device For Stock Price Prediction Using Deep Learning And Explainable AI' Filed by Prof. Pinky Suresh Gangwani; Dr. Vikas Bhowate; Prof. Sachin Subhashrao Patil; Prof. Snehal Awachat; Prof. Pooja Wagh; and Prof. Rihal Ramesh Kalbende

MUMBAI, India, March 13 -- Intellectual Property India has published a patent application (202621009951 A) filed by Prof. Pinky Suresh Gangwani; Dr. Vikas Bhowate; Prof. Sachin Subhashrao Patil; Prof. Snehal Awachat; Prof. Pooja Wagh; and Prof. Rihal Ramesh Kalbende, Nagpur, Maharashtra, on Jan. 30, for 'a method and device for stock price prediction using deep learning and explainable ai.'

Inventor(s) include Prof. Pinky Suresh Gangwani; Dr. Vikas Bhowate; Prof. Sachin Subhashrao Patil; Prof. Snehal Awachat; Prof. Pooja Wagh; and Prof. Rihal Ramesh Kalbende.

The application for the patent was published on March 13, under issue no. 11/2026.

According to the abstract released by the Intellectual Property India: "The invention presents a novel method and system for predicting stock prices by integrating deep learning (DL) with explainable artificial intelligence (XAI). In financial markets, accurate forecasting is essential yet challenging due to volatility and data complexity. The proposed system addresses this by employing DL models, such as LSTM or Transformers, to analyze time-series data including historical prices, technical indicators, sentiment from news/social media, and economic factors. These models capture intricate patterns for precise short- and long-term predictions. To overcome the opacity of DL, XAI techniques like SHAP and LIME are incorporated, generating interpretable explanations (e.g., feature importance scores) that reveal how inputs influence outputs. For instance, a prediction might attribute a price increase to high trading volume (35% impact) and positive sentiment (45%). The system, deployable as a software application or dedicated device, features real-time data processing via APIs, GPU-accelerated computation, and a user-friendly dashboard for visualizations. Novel aspects include the hybrid DL-XAI framework, adaptive model retraining for market shifts, and multi-modal data fusion, enhancing accuracy (up to 95% in tests) over traditional methods. Benefits encompass improved investor confidence, regulatory compliance, and risk management. Applicable to global exchanges like NSE or NYSE, it supports diverse assets. This technical advancement provides a scalable, transparent solution for financial decision-making, bridging prediction power with explainability."

Disclaimer: Curated by HT Syndication.