Detecting Spam Content in Arabic Tweets
Ebtesam Mohammed Al-Qahtani · 2019
The evolution of information has led to an increased intensity in its flow, especially in social communication networks. Twitter, for example, has become an incredibly popular platform for information sharing and opinion expression. Unfortunately, spammers have exploited this situation by promoting their messages and seeking malicious purposes. Various researchers have struggled to tackle this problem, proposing many techniques for the spam detection process. While these studies have made important contributions to the field, they remain limited in their linguistic scope. The current body of literature has focused on English texts with few resources available in the Arabic language. Accordingly, this study proposed an effective method for detecting spam content in Arabic tweets, using a supervised machine learning system. This work employed a set of language-specific features with other features in order to attain a high level of accuracy in the detection process. The proposed approach was evaluated using a real-life dataset and standard evaluation measures. In conclusion, our study shows that the spam content can be detected by using Naive Bayes classifier with accuracy 94%.