MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123030 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 5, 2025, for 'air quality prediction using attention based transformer networks.'
Inventor(s) include Dr. Arivoli A; Sayandeep Chaudhuri; and Aditya Raj.
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: "The present disclosure provides a system for air quality prediction using attention-based transformer networks. The system includes a data preprocessing module configured to receive air quality data comprising concentrations of multiple air pollutants including PM2.5, PM10, CO, NO, NO2, O3, SO2, and NH3, and to normalize the data using robust scaling. The system includes a dual-pathway encoder architecture comprising temporal and spatial embedding pathways that process preprocessed features through learnable embeddings and spatial patches respectively. The architecture includes encoder blocks with multi-head attention mechanisms using rotary positional embeddings RoPE and fully connected networks with swish activation functions. A fusion module combines flattened outputs from both pathways through addition, and an output neural network generates simultaneous predictions for concentrations of the multiple air pollutants."
Disclaimer: Curated by HT Syndication.
