Document Categorization with Entropy Based TF/IDF Classifier

The task of text categorization is assigning a given text document to one or more predefined categories. High availability of digital data requires methods for automatic processing of this data. Day-by day increase of this digital data gives rise to the need of fast and better text classifiers. This...

Full description

Saved in:
Bibliographic Details
Published in2009 WRI Global Congress on Intelligent Systems Vol. 4; pp. 269 - 273
Main Authors Yi-hong Lu, Yan Huang
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2009
Subjects
Availability
Clustering algorithms
Computer science
Educational institutions
Entropy
Information filtering
Information filters
Intelligent systems
Mutual gain information
Mutual information
Text categorization
TFIDF
Online AccessGet full text

Cover

More Information
Summary:The task of text categorization is assigning a given text document to one or more predefined categories. High availability of digital data requires methods for automatic processing of this data. Day-by day increase of this digital data gives rise to the need of fast and better text classifiers. This paper mainly focuses on classifying data in context of text categorization. This paper reports a study conducted on 20 news group dataset, using TFIDF in the context of document categorization. Feature selection is added to this result to improvise the categorization. The results achieved using this algorithm are very promising when compared to conventional methods with features chosen on the basis of bag-of-words text.
ISBN:9780769535715
0769535712
ISSN:2155-6083
2155-6091
DOI:10.1109/GCIS.2009.311