Sentiment Analysis in Social Media Platforms: The Contribution of Social Relationships

The massive amount of data in social media platforms is a key source for companies to analyze customer sentiment and opinions. Many existing sentiment analysis approaches solely rely on textual contents of a sentence (e.g. words) for sentiment identification. Consequently, current sentiment analysis systems are ineffective for analyzing contents in social media because people may use non-standard language (e.g., abbreviations, misspellings, emoticons or multiple languages) in online platforms. Inspired by the attribution theory that is grounded in social psychology, we propose a sentiment analysis framework that considers the social relationships among users and contents. We conduct experiments to compare the proposed approach against the existing approaches on a dataset collected from Facebook. The results indicate that we can more accurately classify sentiment of sentences by utilizing social relationships. The results have important implications for companies to analyze customer opinions.