MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202641094556 A) filed by Cmr College Of Engineering & Technology; Cmr Institute Of Technology; and Cmr Technical Campus on August 04, 2026, for System And Method For Graph Based Context Propagation In Sparse Data Learning Environments.
Inventors include Mr. M. Tanweer Alam; P. Devendra Babu; P Deepthi; K Renuka; Dr Md Shareef; and J Shiva.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present disclosure relates to a system and method for graph-based context propagation in sparse data learning environments. The disclosed system comprises an input data acquisition module configured to receive sparse datasets from one or more heterogeneous data sources, a data preprocessing module configured to normalize the received data and generate feature representations, a feature embedding engine configured to produce latent feature embeddings, and a graph construction engine configured to generate an adaptive contextual graph comprising graph nodes and weighted graph edges. A sparse representation optimization engine optimizes the enriched node representations to preserve contextual diversity and improve learning effectiveness, and a prediction engine generates prediction outputs for one or more downstream machine learning tasks. The disclosed system dynamically updates graph connectivity in response to evolving contextual relationships, enabling reliable context propagation across sparse datasets while reducing propagation errors and improving representation quality.
Disclaimer: Curated by HT Syndication.
