High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Web Document Classification Using Improved Graph Based KNN Classification

Author(s):

Sana Irfan , Bharat Institute of Technology Meerut; Latika Sharma, Bharat Institute of Technology

Keywords:

Mutual Information, Weight of Term, KNN, Feature Selection, Data Mining

Abstract

this paper presents the classification of web document using improved graph based KNN. Most of the organization are facing problem of large amount of unorganized data. Large amount of data is present in the form text journels, email etc. there are many feature selection methods available like Mutual Information, Regularized mutual information, CHI Statics, term Strength, Document Frequency, Inverse Document Frequency, category Term Descriptor, Strong class information word. Feature selection method plays an important role in Text Categorization. The emphasis laid on the combination of standard method localized method i.e weight of term and standard Dataset Reuters-21578 is used to verify the result.

Other Details

Paper ID: IJSRDV3I70056
Published in: Volume : 3, Issue : 7
Publication Date: 01/10/2015
Page(s): 201-203

Article Preview

Download Article