Online product reviews are a major source of business intelligence (BI) that helps managers and market researchers make important decisions on product development and promotion. However, the large volume of online product review data creates significant information overload problems, making it difficult to analyze users’ concerns. In this paper, we employ a design science paradigm to develop a new framework for designing BI systems that correlate the textual content and the numerical ratings of online product reviews. Based on the framework, we developed a prototype for extracting the relationship between the user ratings and their textual comments posted on Amazon.com’s Web site. Two data mining algorithms were implemented to extract automatically decision rules that guide the understanding of the relationship. We report on experimental results of using the prototype to extract rules from online reviews of three products and discuss the managerial implications.
Chung, Wingyan and Tseng, Tzu-Liang (Bill), "Extracting Business Intelligence from Online Product Reviews: An Experiment of Automatic Rule-Induction" (2010). ICIS 2010 Proceedings. 233.