Toward a better fitness club: Evidence from exerciser online rating and review using latent Dirichlet allocation and support vector machine

Abstract

Fitness clubs have never ceased searching for quality improvement opportunities to better serve their exercisers, whereas exercisers have been posting online ratings and reviews regarding fitness clubs. Studied together, the quantitative rating and qualitative review can provide a comprehensive depiction of exercisers’ perception of fitness clubs. However, the typological and dimensional discrepancies of online rating and review have hindered the joint study of the two data sets to fully exploit their business value. To this end, this study bridges the gap by examined 53,979 pairs of exerciser online rating and review from 100 fitness clubs in Shanghai, China. Using latent Dirichlet allocation (LDA) based text mining, we identified the 17 major topics on which the exercisers were writing. A support vector machine (SVM) classifier was then employed to establish the rating-review relations, with an accuracy rate of up to 86%. Finally, the relative impact of each topic on exerciser satisfaction was computed and compared by introducing virtual reviews. The significance of this study is that it systematically creates a standardized protocol of mining and correlating the massive structured/quantitative and unstructured/qualitative data available online, which is readily transferable to the other service and product sectors.

Looking beyond the stars: A description of text mining technique to extract latent dimensions from online product reviews

“…Reviews can provide a depiction of customers’ perceptions of products. Product quality improvement hints can be captured by analyzing users’ online data (Jia, 2019). Scientifically this is important because researchers should not only look at valence and volume of reviews but also examine the content of reviews (Srivastava & Sharma, 2017).…”

“…More recently, authors tried to broaden and deepen the understanding of how reviews can influence product performance. However, the applications of text-mining techniques to examine reviews in more depth has been lacking in the literature, partly because of the complexity of text analysis (Heng et al, 2018; Jia, 2019; Lee et al, 2017; Li, Mai, & Wu, 2018).…”

“…The first contribution of this research is a methodological advancement by providing a technique to disentangle unstructured review texts to identify innovative efforts and sentiment expressed toward product attributes in reviews and check how attributes contribute to higher satisfaction. It provides a protocol for mining and correlating the unstructured data available online, which can be transferred to other industries (Jia, 2019).…”

Managing Customers' Satisfaction: Evidence from Mining User Generated Content on the Internet

“…For this purpose, multilinear regression model was applied to estimate the rating-review relation (Büschken & Allenby, 2016;Jia, 2018a). Moreover, because treating the words in one topic together as a single factor results in coarse grain clusters and thus limited managerial insights (Büschken & Allenby, 2016;Jia, 2018b), the topics were manually broken and reorganised into smaller word groups, with each group representing one fine grain factor. Vol.…”

User Generated Information on Mobile Channels With More Concise Reviews and More Extreme Ratings

“…Online word of mouth has won great attention from marketing scholars over the years [1]- [8]. Online word of mouth is defined as the positive or negative statement addressed by former consumers about a product or service, which is accessible by a large group of people and institutions on the internet [9].…”