MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202641066493 A) filed by Mallesh Babu S; Dr. M. Giri; Dr. V. K. Sharma; Zabi Ur Rahaman K; Mahesha S; and Dr. D. Nagaraju on May 27, 2026, for An Efficient Grid And Density Based Clustering Technique To Detect Global Noisy Samples From High Dimensional Dynamic Dataset.
Inventors include Mallesh Babu S; Dr. M. Giri; Dr. V. K. Sharma; Zabi Ur Rahaman K; Mahesha S; and Dr. D. Nagaraju.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: Clustering methods are used in many applications like psychology, computers, biology, and so technique. DBSCAN algorithm clustering data samples based on density values, but all clustering techniques are proposed with various types of challenges. DBSCAN clustering method is also fail to detect all noisy points. In this research paper we a proposed a new novel efficient DBSCAN (EGDBSCAN) algorithm to cluster data set. First data samples are placed into grid cells, for each cell calculate LNPS, MLNPS, and NNPG values, all new data samples are added one by one to grid space and update the above mentioned parameters, and finally identify list of top 'k' global noisy points. Performance of proposed clustering is compared with KNN noisy detection method and we come to know that proposed algorithm shows better results.
Disclaimer: Curated by HT Syndication.
