6. Conclusions
In this paper, we present a methodology framework to study massive customers’ online “purchasing-reviewing” behaviors. Also, we present a review-to-aspect mapping method to explore the reviewers’ opinions for an online shopping experience in massive and sparse reviews. The analytical and experimental results with real data from a Chinese B2C website of www.jd.com demonstrate that the reviewers grouped by the similar reviewing behavior dynamics can reveal certain information about reviewers’ motivations and concerns for an online shopping experience. This study obtains two major findings:
1) The frequency of time intervals between consumers’ purchasing a good online and their publishing reviews follows a power-law distribution, providing new evidence for the study of human behaviors online. Moreover, similar reviewing behavior dynamics exist for different online shopping experiences in the same B2C website.
2) Different people may use various words to express the same view for an online shopping experience, leading to the sparse distribution of words and increasing difficulty in analyzing customers’ opinions. Review-to-aspect mapping is a feasible method to address this problem. By this method, people can obtain opinions from a group of consumers and then summarize management oriented patterns from these online review data.
