• Naga Sudha
• Y.Madhavee Latha

Abstract- Text classification is become important when the information is increasing rapidly over the internet. This information is in unstructured form and need to be digitized. As these documents are digital form it is necessary for organizing the data by automatically assigning a set of documents into predefined labels based on their content. It mainly depends on the methods that should be used in each phase improves the efficiency of the document classification. In this paper we propose a classification model that supports both the generality and efficiency. It also discusses some of the major issues involved in automatic text classification such as dealing with unstructured text, handling large number of attributes and natural language processing based techniques, dealing with missing metadata and choice of a suitable machine learning technique for training a text classifier. Both are achieved by following the logical sequence of the process of classifying the unstructured text document step by step and efficiency through various methods are proposed. The experimental results over news articles have been validated using statistical measures of accuracy and F-Score. The results have proven that the methods significantly improve the performance.

Index Terms- Text classification, Logistic regression, Naive Bayes classifier, Support Vector machine, Shillloute Coefficient

"An Improved Text classification for Unstructured Document", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.203-209, June 2019, Available :http://www.jetir.org/papers/JETIR1907K31.pdf

"An Improved Text classification for Unstructured Document", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp203-209, June 2019, Available at : http://www.jetir.org/papers/JETIR1907K31.pdf