CAT: Credibility Analysis of Arabic Content on Twitter

Abstract

Data generated on Twitter has become a rich source for various data mining tasks. Those data analysis tasks that are dependent on the tweet semantics, such as sentiment analysis, emotion mining, and rumor detection among others, suffer considerably if the tweet is not credible, not real, or spam. In this paper, we perform an extensive analysis on credibility of Arabic content on Twitter. We also build a classification model (CAT) to automatically predict the credibility of a given Arabic tweet. Of particular originality is the inclusion of features extracted directly or indirectly from the author's profile and timeline. To train and test CAT, we annotated for credibility a data set of 9, 000 Arabic tweets that are topic independent. CAT achieved consistent improvements in predicting the credibility of the tweets when compared to several baselines and when compared to the state-of-the-art approach with an improvement of 21% in weighted average Fmeasure. We also conducted experiments to highlight the importance of the userbased features as opposed to the contentbased features. We conclude our work with a feature reduction experiment that highlights the best indicative features of credibility.

Credibility Detection in Twitter Using Word N-gram Analysis and Supervised Machine Learning Techniques

“…A recent study [6] stated that fake news published on Twitter during the last American presidential elections in 2016, had a significant effect on voters. Several studies have shown that much of the content on Twitter is not credible [7][8][9]. Research by ElBallouli et al [8] found that approximately 40% of the tweets posted per day are not credible tweets.…”

“…Several studies have shown that much of the content on Twitter is not credible [7][8][9]. Research by ElBallouli et al [8] found that approximately 40% of the tweets posted per day are not credible tweets. Moreover, Gupta et al [9] International Journal of Intelligent Engineering and Systems, Vol.…”

“…Classification based approaches classify tweets into credible and not credible based on features extracted from them using machine learning techniques especially supervised techniques [8,[12][13][14][15][17][18][19]. Supervised machine learning techniques require a ground truth that contains a dataset of annotated tweets with the features related to them.…”

“…Sourcebased features consider characteristics of the user such as the number of followers and if the user is verified. Some studies used a combination of content-based and source-based features [8,17,18]. After the feature dataset is built, the next step is to determine the optimal classification algorithm to train them.…”

“…After the feature dataset is built, the next step is to determine the optimal classification algorithm to train them. Decision trees [8,12,19] and support vector machines (SVM) [14,15] are the most popular supervised learning techniques used for classification.…”