Using Latent Class Analysis to Identify Political Behavior of Moroccan Citizens on Social Media
Zakia Acharoui1, Altaf Alaoui2, Mourad Azhari3, Abdallah Abarda4, Badia Ettaki5, Jamal Zerouaoui6
1Acharoui Zakia*, Laboratory of Engineering Sciences and Modeling, Faculty of Sciences, Ibn Tofail University Kenitra, Morocco.
2Altaf Alaoui, Laboratory of Engineering Sciences and Modeling, Faculty of Sciences, Ibn Tofail University Kenitra, Morocco.
3Mourad Azhari, Laboratory of Engineering Sciences and Modeling, Faculty of Sciences, Ibn Tofail University Kenitra, Morocco.
4Abdallah Abarda, Laboratory of Mathematical Modelling and Economic Calculations, FSJES, University Hassan 1er, Settat, Morocco.
5Badia Ettaki, Laboratory of Research in Computer Science, Data Sciences and Knowledge Engineering, Department of Data, Content and knowledge Engineering School of Information Sciences, Rabat, Morocco.
6Jamal Zerouaoui, Laboratory of Engineering Sciences and Modeling, Faculty of Sciences, Ibn Tofail University Kenitra, Morocco.
Manuscript received on March 15, 2020. | Revised Manuscript received on April 01, 2020. | Manuscript published on April 10, 2020. | PP: 1273-1282 | Volume-9 Issue-6, April 2020. | Retrieval Number: F4095049620/2020©BEIESP | DOI: 10.35940/ijitee.F4095.049620
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The high use of social media has led to a new form of political involvement and participation. In this paper, we use Latent Class Analysis to identify participants’ behavior regarding political participation and engagement based on the nature of their interaction on social media. The LCA findings reveal three statistically distinct and behavioral classes regarding political interaction on social media. The profiles were ranged from ‘Activist’ that show more engagement in political activity, such as following candidates and political parties, posting and participating in discussions related to economic, social or political issues or, encouraging others to debate their point of view, to ‘Agitator’ and ‘Outsider’ profiles that show a low probability of interacting on social media and engaging in political actions. The LCA technique has provided meaningful and distinct information on the participants’ political profile than clustering classical techniques.
Keywords: Latent Class Model, Clustering, Social Media, Political Commitment.
Scope of the Article: Clustering