MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123378 A) filed by Mr. M Balasubramanian; Mr. A Mani; and S. A. Engineering College, Chennai, Tamil Nadu, on Dec. 8, 2025, for 'machine learning interpretability stellarbasin.'

Inventor(s) include Mr. M Balasubramanian; and Mr. A Mani.

The application for the patent was published on Jan. 2, under issue no. 01/2026.

According to the abstract released by the Intellectual Property India: "MACHINE LEARNING INTERPRETABILITY STELLARBASIN STELLARBASIN is an interactive web-based visualization platform designed to provide intuitive insights into pre-trained machine learning models. The primary objective of the system is to bridge the gap between machine learning complexity and interpretability by allowing users to upload trained models and receive clear, dynamic visual representations along with detailed model reports. This approach aims to aid data scientists, educators, and analysts in understanding model behaviour without delving into the underlying training process. The platform supports common supervised learning models, including Decision Trees, K-Nearest Neighbours (KNN), Linear Regression, and Logistic Regression. Upon uploading a model, users receive visual explanations tailored to the specific algorithm - such as decision tree structures, feature influence plots, and regression lines - alongside model summaries like accuracy, coefficients, and class probabilities. STELLARBASIN is developed using a modular architecture: the frontend is built with React.js for an interactive and responsive user experience, while the backend combines Flask and Dash for model processing and rendering visualizations. Core libraries include scikit-learn, matplotlib, seaborn, and plotly, ensuring robust analysis and graphical outputs. With built-in sample data input and validation, the platform also enables users to simulate predictions and observe outcomes in real time. Future enhancements may include support for more complex models, comparative analysis, and automated explainability tools, further expanding the system's utility in model interpretation and education."

Disclaimer: Curated by HT Syndication.