Characterizing comment spam in the blogosphere through content analysis

Abstract

Abstract-Spams are no longer limited to emails and webpages. The increasing penetration of spam in the form of comments in blogs and social networks has started becoming a nuisance and potential threat. In this work, we explore the challenges posed by this type of spam in the blogosphere with substantial generalization regarding other social media. Thus, we investigate the characteristics of comment spam in blogs based on their content. The framework uses some of the previously explored methods developed to effectively extract the features of the blog spam and also introduces a novel method of active learning from the raw data without requiring training instances. This makes the approach more flexible and realistic for such applications. We also incorporate the concept of cotraining for supervised learning to get accurate results. The preliminary evaluation of the proposed framework shows promising results.

A Self-Supervised Approach to Comment Spam Detection Based on Content Analysis

“…This work is inspired and extended the work done by Mishne and Carmel [16] and our previous work [42]. The prototype system has appeared to be more flexible than most supervised spam detection systems which depend primarily on pre-classified data.…”

“…This process is explained in more detail in our previous work [42]. This method provides us with a set of potential spam comments.…”

“…In our previous work [42], we compared different algorithms such as Naïve Bayes, Support vector machines (SVM), neural networks, Logistic Regression, Decision Trees etc [42]. The comparison indicated that decision tree (J48) performed the best for the dataset used.…”

“…A higher value of spam similarity will enforce the system to be more precise while reducing recall whereas a modest value of spam similarity will have a high recall system with a compromise in precision. The tradeoff of precision and recall with changing spam similarity is illustrated in our previous work [42]. Since this step aims to generate a training corpus for our second step, we chose a high threshold value to classify comments.…”

A Study of Spam Detection Algorithm on Social Media Networks

“…Archana et al (2009) has dealt with the spam that gets penetrated in the form of comments in Blog. A blog is a type of web content which contains a sequence of periodic user comments and opinions for a particular topic.…”