MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641110405 A) filed by Vellore Institute Of Technology on September 15, 2026, for A Hybrid Quantum-Classical Clustering System Using Entropy-Weighted Quantum Encoding And Quantum Optimization.
Inventors include M. Parimala; and Naresh L.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention relates to a hybrid quantum-classical clustering system and method for unsupervised data clustering. The system comprises a data preprocessing module configured to normalize multidimensional feature vectors, an entropy weighting module configured to determine entropy-derived feature weights, and a quantum encoding module configured to generate amplitude-encoded quantum states from entropy-weighted feature vectors. A quantum similarity evaluation module executes a quantum Swap Test circuit to determine fidelity-based similarity between quantum-encoded data samples and cluster centroids. Preliminary cluster assignments are refined using a Quantum Approximate Optimization Algorithm (QAOA), followed by classical centroid updating and iterative convergence evaluation. The iterative processing continues until a predetermined convergence criterion is satisfied, whereupon final cluster assignments and centroid information are generated. The disclosed architecture integrates entropy-guided feature representation, quantum similarity evaluation, quantum optimization, and classical centroid refinement within a unified hybrid computational framework for processing multidimensional datasets
Disclaimer: Curated by HT Syndication.
