White Papers

Naive Bayes Classifier Learning With Feature Selection for Spam Detection in Social Bookmarking

Overview Social bookmarking systems such as BibSonomy and del.icio.us have become increasingly popular with the prevalent use of internet. These systems provide powerful infrastructure solutions for semantic annotation and information sharing, promoting diverse kinds of internet-based activities, e.g., web exploration, creating and joining web-based communities, and buying recently published volumes. This paper proposes a machine learning-based approach to automatic spam detection. In specific, a set of relevant features, i.e., the number of posts and posted tags for each user are extracted from training data. The extracted tags are sorted by mutual information. Then, the tags, having high mutual information value and used in test data, are chosen for the classification task.

Download White Paper

By downloading you agree to our Terms and Conditions. These include information regarding use of your personal data.

Publisher
Soongsil University
File Format
PDF
Date Published
Nov 29, 2008
Format
White Papers
Topics
Artificial Intelligence, Spam - E-mail Fraud - Phishing

Similiar White Papers

Semi-Supervised Learning: A Comparative Study for Web Spam and Telephone User Churn

Semi-Supervised Learning: A Comparative Study for Web Spam and Telephone User Churn

This paper compares a wide range of semi-supervised learning techniques both for Web spam filtering and for telephone us

Publisher: MTA SZTAKI  |  Tags: network, social network, spam