Use of Semantic Relation Between Words in Text Clustering

In traditional document clustering methods, a document is considered a bag of words. The fact that the words may be semantically relateda crucial information for clusteringis not taken into account. The feature vector representing the document is constructed from the frequency count of document terms. To improve results, weights calculated from techniques like Inverse Document Frequency (IDF) and Information Gain (IG) are applied to the frequency count. These weights also are essentially statistical parameters and do not make use of any semantic information.