Abstract

Clustering is an unsupervised machine learning technique that serves a gargantuan task in passing on the data sets into precise clusters depending on various convergence or divergence characteristics. It has a brawny prospective in health-related data analysis for programmed disease prophecy. K-means is a clustering scheme that is extensively used in various areas of machine learning. The objective of our paper is to upgrade an existing clustering algorithm, K-Mean. The model will be trained using Microarray datasets and the testing will be done using WEKA, this is an open source application. Apparently, from innumerable biological experiments and various community researches, there has been upsurge in the amount and complexity of Micro-array datasets. A storehouse that contains Micro-array gene manifestation data is called a Micro-array database.