Identifying Key Product Attributes and Their Importance Levels From Online Customer Reviews

Abstract

Identifying customer needs and preferences is one of the most important tasks in design process. Typically, a variation of interview based approaches is used to conduct need and preference analysis. In this paper, a new approach based on text mining online (internet based) customer reviews to supplement traditional methods of need and preference analysis is considered. The key idea underlying the proposed approach is to partition online customer generated product reviews into segments that evaluate the individual attributes of a product (e.g zoom capability and support of different image formats in a camcorder). Additionally, the proposed method also identifies the importance (ranking) that customers place on each product attributes. The method is demonstrated on 100 customer reviews submitted for camcorders on epinions.com over a two year period.