Arabic Tweets Spam Detection Based on Various Supervised Machine Learning and Deep Learning Classifiers

Abstract

In this paper, different machine learning algorithms, ensemble algorithms, and deep learning algorithms are applied to Arabic tweets to detect whether it human-generated or not. The tweets are used twice as preprocessed and non-preprocessed to measure the effectiveness of Arabic preprocessing in the classification process. The data is also tokenized with various methods like unigram, trigram, and Term Frequency–Inverse Document Frequency. The experiments show that the support vector machine with the non-preprocessed tweets and unigram tokenization has the best performance of 83.11% and a precision of 0.9516 while it predicts the spam or not in a relatively small time.