Comparison of Supervised and Semi- Supervised Fuzzy Clusters in Text Categorization

Electronics gadgets are part of human life in these days, as a result abundant data is generated and it is growing in exponential rate. Data Generated was earlier stored in dumped repositories. The paper attempts in proposing a classified repository so that at later retrieval of stored data or navigation becomes easy. In the present paper comparison between supervised and semi-supervised classification is explored. Semi-Supervised Classification is half way between the supervised and unsupervised paradigm which can be employed where we have a limited amount of training data. In the process of feature reduction, the features which are words in Text Classification, clusters are formed, that have fuzzy concepts. Three different types of clusters are formed namely soft, hard and mixed and different similarity measures are employed in the process of text classification, the results obtained are encouraging.