31-10-2014, 03:22 PM
Abstracts: In the real-world, the ability to automatically classify documents into a ?xed set of categories which is highly required in so many areas. Common scenarios include classifying a large amount of unclassi?ed documents such as newspaper articles. Newspaper articles can be classi?ed as ’features’, ’sports’ or ’news’ etc. which is done manually in the news paper publication by the knowledgeable person who knows about all the categories. Manual categorization is very time consuming and effort asking. Our proposed system will automatically categorize the articles into their particular categories which reduce the time and manual work of the person. This System broadly classified into 2 phases : 1) Training phase in which the known data set of each category is predefined.2) Testing phase in which the article will be categorized automatically Even if the precise keywords are not the part of articles but semantically related words are there in article. This system will be very useful for English news paper publication.