MUMBAI, India, June 16 -- Intellectual Property India has published a patent application (202611053712 A) filed by Mr. Vikas Choudhary; Mr. Shubham Maheshwari; Mr. Siddhant Gogia; and Mr. Vivek Bansal, Ghaziabad, Uttar Pradesh, on April 27, for 'social influence prediction using graph neural networks.'
Inventor(s) include Mr. Vikas Choudhary; Mr. Shubham Maheshwari; Mr. Siddhant Gogia; and Mr. Vivek Bansal.
The application for the patent was published on June 5, under issue no. 23/2026.
According to the abstract released by the Intellectual Property India: "In recent years, social media platforms have devel- oped in a meaningful environment for the exchange of infor- mation, formation of opinion, and influence among the users. Finding influential users in these large networks is now important for applications like content recommendation, viral content marketing and misinformation control. Traditional graph-based techniques like PageRank and degree centrality are unable to capture complex, non linear relationships between users. To overcome the limitation, this study proposes a framework for social influence prediction i.e based on Graph Neural Network (GNN). The model represents users and their interactions as nodes and edges in a graph and learns latent representation to dynamically compute the influence scores. A prototype was developed using ReactJS for real-time representation of influence spread and top user rankings, PyTorch for the GNN model, and Fast API for backend processing. Testing on the simulated graph datasets represent that the recommended approach performs better than traditional techniques in terms of comprehensible network insights and influence estimation accuracy. Keywords: Influence Prediction, Social Network Analysis, Graph Neural Networks (GNN), Fast API, ReactJS, PageRank, Degree Centrality and PyTorch."
Disclaimer: Curated by HT Syndication.
