Fuzzy Logic Based Dynamic Plotting of Mood Swings from Tweets

摘要

Twitter is one the most popular social media platforms. Users express their feelings easily on social media regarding any trending event. In this paper, we propose a fuzzy logic based approach for dynamic plotting of mood swings from tweets. The novelty of the paper is use of linguistic hedges with fuzzy logic to compute the sentiment of tweet. Comparison of our approach with existing methods, on real-time tweets extracted from online website confirms the superiority and efficiency of our method. The tweets used in our experiments are extracted from the timeline of the India Vs Pakistan final ICC world-cup match in June 2017. They reflect the moods of the twitter users as the match progresses. Using our fuzzy logic based approach, we successfully plot the dynamic mood vs time and compute the polarity of the sentiment at each time instant.