Comparative Performance of Machine Learning and Deep Learning Algorithms for Arabic Hate Speech Detection in OSNs Ahmed Omar 1(&) , Tarek M. Mahmoud 1 , and Tarek Abd-El-Hafeez 1,2 1 Computer Science Department, Faculty of Science, Minia University, EL-Minia, Egypt {Ahmed.omar,d.tarek,tarek}@mu.edu.eg 2 Deraya University, EL-Minia, Egypt Abstract. Nowadays, Online Social Networks (OSNs) are the most popular and interactive media that used to express feelings, communicate and share information between people. However, along with useful and interesting con- tent, sometimes unsuitable or abusive content can be published on these net- works, such as hate speech and insults. Hate speech includes any type of online abuse concepts like cyberbullying, discrimination, abusive language, profanity, flaming, toxicity, and harassment. Most of the Hate speech detection attempts have concentrated on the English text, while work on the Arabic text is sparse. In this paper, we constructed a standard Arabic dataset that can be used for hate speech and abuse detection. In contrast to most previous work the datasets were collected from one platform, the proposed dataset is collected from more social network platforms (Facebook, Twitter, Instagram, and YouTube). To validate the effectiveness of the proposed datasets twelve machine learning algorithms and two deep learning architecture were used. Recurrent Neural Network (RNN) outperformed other classifiers with an accuracy of 98.7%. Keywords: Arabic hate speech Á Hate speech detection Á Arabic text classification Á OSN 1 Introduction The rapid development of online social networks (OSNs) led to the widening of the scope of the internet. Nowadays OSN is the most popular and interactive media that used to express feelings, communicate and share information between people. The communication includes all data types such as text, image, audio, and video. Text is the most exchanged content type on social media via posts, tweets, comments, replies, and messages. Facebook is the top social network site with 2.414 billion active users. YouTube, as a video streaming site, ranked second with 2 billion active users, Instagram, as a photo- sharing site, ranked sixth with 1 billion active users. Despite his widespread reputation, Twitter ranked eleventh with 300,000 active users [6]. Every minute, 510,000 com- ments, and 293,000 status updates are posted on Facebook, 4.15 Million videos are © Springer Nature Switzerland AG 2020 A.-E. Hassanien et al. (Eds.): AICV 2020, AISC 1153, pp. 247–257, 2020. https://doi.org/10.1007/978-3-030-44289-7_24