Social analytics: Learning fuzzy product ontologies for aspect-oriented sentiment analysis

• DOI: 10.1016/j.dss.2014.05.005
• Publication Date: 2014-05-15T07:32:13Z
• Authors: Raymond Y.K. Lau; Chunping Li; Stephen S.Y. Liao
• Keywords: Social media analytics; Sentiment Analysis; Business Intelligence; Leverage (statistics); Business analytics

In the era of Web 2.0, there has been an explosive growth of consumer-contributed comments at social media and electronic commerce Web sites. Applying state-of-the-art social analytics methodology to analyze the sentiments embedded in these consumer comments facilitates both firms' product design strategies and individual consumers' comparison shopping. However, existing social analytics methods often adopt coarse-grained and context-free sentiment analysis approaches. Consequently, these methods may not be effective enough to support firms and consumers' demands of fine-grained extraction of market intelligence from social media. Guided by the design science research methodology, the main contribution of our research is the design of a novel social analytics methodology that can leverage the sheer volume of consumer reviews archived at social media sites to perform a fine-grained extraction of market intelligence. More specifically, the proposed social analytics methodology is underpinned by a novel semi-supervised fuzzy product ontology mining algorithm. Evaluated based on real-world social media data, our prototype system shows remarkable performance improvement over a baseline ontology learning system and a context-free sentiment analysis system. The managerial implication of our research is that firms can apply the proposed social analytics methodology to tap into the collective social intelligence on the Web, and hence improve their product design and marketing strategies.