With the explosive growth of social media on web, analyzing Public Sentiment Variations (PSV) has become utmost necessity as public sentiments fluctuate with alterations in real life events. Analysis of PSV empowers decision makers to gain a better understanding of public reactions in social, political and economic environment thus helps in better decision making. Most of the recent studies are bounded to be analyzing and predicting public sentiments. In this paper, we have done further analysis to know the useful insights for PSV using tweets and Facebook comments with a specific target. We propose a Novel Reason Mining Algorithm to find the possible reasons affecting PSV in significant sentiment variation period. It uses incremental iterative approach to refine the final list of most influential reasons. To give more intuitive representation, the algorithm ranks a set of mined reason candidates. The reason candidate with the most number of tweets/comments is the main reason for sentiment variation in that time period. The quality of mined reasons mainly depends on the quality of clusters formed prior to reason mining. The performance of our approach is compared with baseline K-means clustering technique. Experimental results show that our approach forms better clusters and effectively mine the reasons for sentiment variations. The Kmeans technique attained 57.05% precision and 53.31% recall on Twitter and 69.86% precision and 40.62% recall on Facebook. The Novel Reason Mining Algorithm achieved 79.4 % precision and 44.45% recall on Twitter and 71.13 % precision and 40.62% recall on Facebook.24 possible reasons behind sentiment variations. METODOLOGYThe steps performed for analyzing public sentiment variations and mining the useful insights behind these variations are described in this section. First, the tweets and Facebook comments regarding specific target (e. g. "Narendra Modi") are extracted from the collected tweets and comments. The
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.