Discriminative Features Selection in Text Mining Using TF-IDF Scheme
This paper describes technique for discriminative features selection in Text mining. 'Text mining' is the discovery of new, previously unknown information, by computer. Discriminative features are the most important keywords or terms inside document collection which describe the informative news included in the document collection. Generated keyword set are used to discover Association Rules amongst keywords labeling the document. For feature extraction Information Retrieval Scheme i.e. TF-IDF is used. This system uses previous work, which contains Text Preprocessing Phases (filtration and stemming).