Abstract

To store the textual data and various documents, the usage of electronic media is widespread. To retrieve theimportant information from the large document collection of unstructured data is a very difficult and time consuming task. Itis easier to find the relevant documents from a huge data collection only when the data collected is in ordered form or thedata is classified by certain group or category. Still the problem persists to find the best grouping technique. This paperconcentrates on the implementation technique of k-means clustering algorithm. K-means technique is used here to cluster theunlabeled data or the text document collection that is highly unstructured. It begins with the representative model of theunstructured data and finally generating the set of sorted clusters as a result. Furthermore, the results can be refined byanalyzing the sorted set of clusters.
Keywords: K Means, Vector Space Model (VSM), Euclidean Distance, Text Clustering, Residual sum of square, TF-IDF, HAC.