PurposeThis work aims to explore why people review their experienced online shopping in such a manner (promptness), and what is the potential relationship between the users’ review promptness and review motivation as well as reviewed contents.Design/methodology/approachTo evaluate the customers’ responses regarding their shopping experiences, in this paper, the “purchase-review” promptness is studied to explore the temporal characteristics of users’ reviewing behavior online. Then, an aspect mining method was introduced for assessment of review text. Finally, a theoretical model is proposed to analyze how the customers’ reviews were formed.FindingsFirst, the length of time elapsed between purchase and review was found to follow a power-law distribution, which characterizes an important number of human behaviors. Within online review behaviors, this meant that a high frequency population of reviewers tended to publish relatively quick reviews online. This showed that the customers’ reviewing behaviors on e-commerce websites may have been affected by extrinsic motivations, intrinsic motivations or both. Second, the proposed review-to-feature mapping technique is a feasible method for exploring reviewers’ opinions in both massive and sparse reviews. Finally, the customers’ reviewing behaviors were found to be mostly consistent with reviewers’ motivations.Originality/valueFirst, the authors propose that the “promptness” of users in posting online reviews is an important external manifestation of their motivation, product experience and service experience. Second, a semi-supervised method of review-to-aspect mapping is used to solve the data quality problem in mining information from massive text data, which vary in length, detail and quality. Finally, a huge amount of e-commerce customers’ purchase-review promptness are studied and the results indicate that not all product features are responsible for the “prompt” posting of users’ reviews, and that the platform’s strategy to encourage users to post reviews will not work in the long term.

In this work, we propose a research framework to exploring the useful information about airline service from massive online reviews, especially, the airline service features from the customer perspective. The experimental results indicate that the proposed methods can extract information about customers' opinion about the airline service features. The common concerns as well as special features for different airline company can also be extracted efficiently from the massive online review data.

No abstract

No abstract