SUBJECTIVITY AND SENTIMENT ANALYSIS OF AMHARIC COMMENTS ON SOCIAL MEDIA: THE CASE OF ETHIOPIA POLITICAL DISCOURSE

Abstract

The rise of social media and the rapid development of mobile communication technologies have dramatically changed the way to express the feeling, attitude, lifestyle, opinions, events, situations (like based on political, economic, and social point of view), passion, etc. People often express these behaviors through social media in the form of short texts. Concerning events and situations on behalf of the political sector, social media can be an enabler for participation and democracy among citizens of participating groups. This study is conducted to perform and gain insight into political sentiment using machine learning technique from the comment given in Amharic language on the topic of the idea which is posted through social media so as to understand the subjectivity of the comment. The study try to demonstrate the analysis of the sentiments and it’s a subjectivity to identify their trends, numerical implications. The data that is considered in this research is based on two data sets that are self-build, The first one is a data set for the subjectivity of the comment and the other one is for sentiment analysis getting after subjective sentence or comments that are extracted but the main content of data set is the same. Finally, the results show the trends in social media users comment sentiments and subjectivity correlation to the real-world events associated with the respective keywords and provide a clear picture of the influence of real-world events on social media user’s comment and it’s sentiments of polarity.