Smart Learning in Document Categorization using Dynamic Learning

Abstract

Clustering is the process of making data groups using similar data items, used for data mining to extract data from available large datasets. A large volume of text documents consisting of personal information is being generated in form of digital libraries and repositories in internet daily.It is conceivable to get to great quality instructive substance and strategies in an increasingly helpful manner. In spite of the fact that a ton of keen instruments have been connected for instructive application, there are just restricted looks into that show the instructive viability of shrewd devices through test contemplations, Clustering organizes large quantity of unordered text documents into small number of meaningful and coherent clusters. A clustering method based on K-Means algorithm is proposed in this paper. K-Means is a unsupervised algorithm based on randomly selected initial centroids used to cluster a highly unstructured and unlabeled document collection. The system will be evaluated using precision as a measure.