A proposed spam detection approach for Arabic social networks content
M’hamed Mataoui, Omar Zelmati, Dalila Boughaci, Moncef Chaouche, Fatima Lagoug · 2017
Data collected from social networks can be used to discover useful information by mining users' opinions for different knowledge fields, such as marketing, politics or social studies. Nowadays, online social networks and reviews platforms become a prime target to different types of malicious and opportunistic actions by creating spam content which can skew the results of an automatic study on user opinion trends. We propose in this paper a spam detection system to process Arabic content generated on social networks. Our approach is based on a set of selected features which characterize Arabic spam content. Our initial tests by using WEKA software show the effectiveness (91.73 % of precision) of our approach and the accurate choice of features.