A comparative study of machine learning techniques in blog comments spam filtering

In this paper we compare four machine learning techniques for blog comments spam filtering. the machine learning techniques are the Naïve Bayes, K-nearest neighbor, neural networks and the support vector machines. For this comparative study we used a blog comment corpus that has been affected by spam, which is our study case in this work. We classify the comments of this blog comments corpus, which have 50 pages and 1024 blog comments are classified in spam an non-spam. The percentage of spam of this corpus is 67%.