The important aspect of automatically sorting and classifying a set of documents into any category by incorporating a predefined set is Text categorization. Automated text classification is gaining notability since it frees organizations form the hectic and time consuming need of manually organizing documents, which can be too expensive, or simply not feasible given the time constraints of the application or the number of documents involved. In terms of accuracy, modern text classification systems proves better than that of trained human professionals, which is made possible by a combination of information retrieval technology and machine learning technology in text classification approach. There are numerable useful application of this approach spanning various scientific and general fields of work. This paper deals in dept the feasibility of text categorization pertaining to various domains along with making substantial use of techniques like document indexing, text filtering and classifier learning technique. Also the approaches of standard input and tokenization are considered for a better out which shall be devoid of any complexity for text classification.