Bayesian methods for classification inappropriate web pages

The incursion of the Internet has created new forms of information and communication, but it can also carry great dangers, when its use is related to inappropriate content, such as, access to harmful contents and the rise of new kinds of crimes.   In this situation, automatic filtering systems ident...

Full description

Saved in:
Bibliographic Details
Published inVisión Electrónica Vol. 11; no. 2; pp. 179 - 189
Main Authors Rodríguez R, Jorge E., Ortiz Pimiento, Jenny Paola
Format Journal Article
LanguageEnglish
Published Universidad Distrital Francisco José de Caldas 27.10.2018
Subjects
bayesian methods
data preprocessing
machine learning
web content filtering
web mining
Online AccessGet full text

Cover

More Information
Summary:The incursion of the Internet has created new forms of information and communication, but it can also carry great dangers, when its use is related to inappropriate content, such as, access to harmful contents and the rise of new kinds of crimes.   In this situation, automatic filtering systems identify improper Internet content. This paper describes the use of an algorithm, to automatically filter out inappropriate Web pages. To accomplish this (automatic filtering) task implementation method TAN (Tree Augmented Naive Bayes) is plasma. Data mining algorithms and computational learning for the extraction process, representation and classification of web pages are implemented.
ISSN:1909-9746
2248-4728
DOI:10.14483/22484728.14626