MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202521131041 A) filed by Symbiosis International, Pune, Maharashtra, on Dec. 23, 2025, for 'system and method for vehicle transaction analysis using gradient boosting with native categorical processing.'
Inventor(s) include Bhupesh Kumar Dewangan; and Pawan Hete.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "A computer-implemented system for analyzing vehicle transaction data using machine learning to predict transaction parameters and classify vehicle characteristics is disclosed. The system comprises a processor and a memory storing executable instructions. The system includes a data acquisition module receiving vehicle transaction data with high-cardinality categorical features, a data preprocessing module grouping rare categories and converting features to category data type eliminating one-hot encoding requirements, a feature engineering module applying logarithmic transformation to skewed count targets and decomposing dates into temporal components, a machine learning engine training XGBoost models with native categorical support and GPU acceleration to generate regression and classification models, a prediction module generating transaction count predictions and fuel type classifications through the trained models, and an output interface module transmitting results with confidence metrics to user devices for visualization. The system processes million-record datasets without memory overflow, achieves R-squared exceeding zero point six for regression."
Disclaimer: Curated by HT Syndication.
