A Spam Filtering Method Learning from Web Browsing Behavior
نویسندگان
چکیده
In this paper a spam filtering method is proposed. We focus on user behavior that most email users browse the Web. The method reduces troublesome maintenance of the spam filter, since the filter learns from Web browsing behavior in the background. The method uses Web browsing behavior of each user to learn ham words. Ham words are picked up from browsed Web pages using TF-IDF and stored in the database called ham words list. For each received email, the method extracts keywords from the email, including Web pages of the URLs. If some keywords are in the ham words list, the email is treated as a ham. In our experiments, several spam emails which cannot be detected by a Bayesian filter are detected as spams.
منابع مشابه
SNUMedinfo at TREC Web track 2014
This paper describes the participation of the SNUMedinfo team at the TREC Web track 2014. This is the first time we participate in the Web track. Rather than applying more sophisticated retrieval method such as learning to rank models, this year we used only baseline retrieval models with spam filtering and pagerank prior.
متن کاملWeb Spam Classification: a Few Features Worth More
In this paper we investigate how much various classes of Web spam features, some requiring very high computational effort, add to the classification accuracy. We realize that advances in machine learning, an area that has received less attention in the adversarial IR community, yields more improvement than new features and result in low cost yet accurate spam filters. Our original contributions...
متن کاملLearning Spam: Simple Techniques For Freely-Available Software
The problem of automatically filtering out spam e-mail using a classifier based on machine learning methods is of great recent interest. This paper gives an introduction to machine learning methods for spam filtering, reviewing some of the relevant ideas and work in the open source community. An overview of several feature detection and machine learning techniques for spam filtering is given. T...
متن کاملSemi-Supervised Learning: A Comparative Study for Web Spam and Telephone User Churn
We compare a wide range of semi-supervised learning techniques both for Web spam filtering and for telephone user churn classification. Semisupervised learning has the assumption that the label of a node in a graph is similar to those of its neighbors. In this paper we measure this phenomenon both for Web spam and telco churn. We conclude that spam is often linked to spam while honest pages are...
متن کاملGraph Labelling Workshop and Web Spam Challenge
We compare a wide range of semi-supervised learning techniques both for Web spam filtering and for telephone user churn classification. Semisupervised learning has the assumption that the label of a node in a graph is similar to those of its neighbors. In this paper we measure this phenomenon both for Web spam and telco churn. We conclude that spam is often linked to spam while honest pages are...
متن کامل