Research Spotlight

MIS Faculty:

Research Interests:

• Business intelligence
• Information retrieval
• Natural language processing
• Text/data mining
• Machine learning

Research Title: "Controversy is Marketing: Mining Sentiments in Social Media."

Publication Information: In the 43rd Hawaii International Conference on System Sciences (HICSS) 2010.

Abstract: Is there any business value in consumer-generated product reviews? How are consumer opinions correlated with product sales? Does a product sell well when consumer opinions converge or diverge? This research attempts to answer these questions.

Enabled by Web 2.0 technologies, social media provide an unparalleled platform for consumers to share their product experiences and opinions—i.e., through word-of-mouth (WOM) or consumer reviews. It has become increasingly important to understand how WOM content and metrics of consumer review texts influence consumer purchases and product sales.

By integrating marketing theories with text-mining techniques, we propose a set of novel measures that focus on sentiment divergence in consumer product reviews. To test the validity of these metrics, we conduct an empirical study based on data from Amazon.com and bn.com (Barnes and Noble). Our results demonstrate a strong causal relationship between our proposed measures and product sales. This effect is not fully captured by the non-textual review measures. Furthermore, our divergence metrics are shown to be superior and more appropriate to capturing the sales effect of review content than some commonly used textual measures in the literature.

The findings provide important insights into the business impact of social media and user-generated content, an emerging problem in business intelligence research. From a managerial perspective, our results suggest that, when managing social media, firms should pay special attention to the textual content information, and more importantly focus on the right measures. [full text]

For additional information, please contact us.