MULTI-LEVEL ANALYSIS OF POLITICAL SENTIMENTS USING TWITTER DATA: A CASE STUDY OF THE PALESTINIAN-ISRAELI CONFLICT

Given the tremendous growth of social media platforms, people have been actively spreading not only information in general, but also political opinions. Many research efforts have used social media content to analyse and predict the public opinion towards political events. This work presents an analytical study for measuring the political public opinion towards the Palestinian-Israeli conflict by using Twitter data. The study uses a novel data analysis model that leverages two levels of analysis: country-level analysis and individual-level analysis. The country-level analysis aims to explore the country s overall attitude towards Palestine by: 1) Identifying counties that generated the most topic-focused tweets, 2) Measuring the friendliness of each country towards Palestine. 3) Analysing the change of sentiment over time. The individual-level analysis aims to analyse data based on the activity and background of individuals. The attitudes of opinion leaders and ethnic groups were analysed and discussed in light of countries attitudes. The rich experience provided in this study through the proposed model for multi-level analysis, the step-by-step procedure, the variety of analysis techniques and the discussion of results can be informative for other developers and data analysts who are interested in analysing social media sentiment about political conflicts in particular.