Comparison of series products from customer online concerns for competitive intelligence

Abstract

Online reviews provide valuable information for product designers and the integration of online concerns into new product design has been investigated by different researchers. However, few of them exploit the value of online concerns on the comparison of series products. Analyzing online concerns of series products facilitates designers to obtain shared customer preferences regarding products in a series and recognize the strength and weakness of products in competitive series. Accordingly, a framework is designed to discover shared pros and cons of series products by exploring online customer concerns, in which representative opinionated sentences are sampled from reviews of series products. In particular, opinionated sentences of specific features are initially identified from product reviews. Then, opinionated sentences regarding the same series products are clustered, which helps to extract similar customer concerns. Finally, an optimization problem is formulated for the sampling of a few opinionated representative sentences. With a large number of real data from Amazon.com, categories of experiments were conducted to evaluate the effectiveness of the proposed approach. This study explores to integrate big consumer data for competitive intelligence in the market driven new product design, which helps the theoretical development on customer requirement management in the fierce market.