INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Using Machine Learning Algorithms For Analysis Of Spam And Its Detection
Authors Name:
Kumar Jayantilal Parmar
, Chintan B. Thacker
Unique Id:
IJSDR1610032
Published In:
Volume 1 Issue 10, October-2016
Abstract:
Web spam is one of the major problems of search engines because it reduces the quality of the Web page. Web spam also effects economically because spammers provide a large free advertising data or sites on the search engines and so an increase in the web traffic. There are certain ways to distinguish such spam pages and one of them is using classification techniques. Comparative analysis of web spam detection using machine learning algorithm like LAD Tree, and Random Forest, C4.5 and Naive bayes have been presented in this paper. Experiments were carried out on feature sets of universally accepted dataset WEB SPAM UK-2007 using WEKA. By observing all the results we found that Random forest works well on content based features, link based features and transformed link based features. But few techniques were found time consuming as compared to other classification techniques used.
Keywords:
Machine learning, Spamdexing, cloaking, link spam, content spam, C4.5, Naive bayes, LAD tree, decision tree, Random forest, Web search engine, attribute selection.
Cite Article:
"Using Machine Learning Algorithms For Analysis Of Spam And Its Detection", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.1, Issue 10, page no.191 - 194, October-2016, Available :http://www.ijsdr.org/papers/IJSDR1610032.pdf
Downloads:
000337064
Publication Details:
Published Paper ID: IJSDR1610032
Registration ID:160882
Published In: Volume 1 Issue 10, October-2016
DOI (Digital Object Identifier):
Page No: 191 - 194
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn