Start Date
11-12-2016 12:00 AM
Description
In online markets, with the convenient and extensive information search, conventional classification methods cannot afford a precise understanding of products. This research draws on comprehension research and posits that the perceptual schema used by consumers to comprehend product information varies for different products. As product reviews are a major source of product-related information, we use product review perception to derive the perceptual schemas. In the paper, we present our three-step method in detail and use it to generate preliminary product clusters. As an exploration of product classification, this research contributes in several perspectives. First, our generated clusters help understand consumer behaviors towards different products. Second, we provide schema prototypes which depict consumers’ perceptual sets towards different products, contributing to both research and practices of online markets. Third, instead of a top-down approach of classifying products, our bottom-up method provides insights of using and mining the value of online textual content.
Recommended Citation
Bao, Zhuolan and Chau, Michael, "A Schema-oriented Product Clustering Method Using Online Product Reviews" (2016). ICIS 2016 Proceedings. 17.
https://aisel.aisnet.org/icis2016/DataScience/Presentations/17
A Schema-oriented Product Clustering Method Using Online Product Reviews
In online markets, with the convenient and extensive information search, conventional classification methods cannot afford a precise understanding of products. This research draws on comprehension research and posits that the perceptual schema used by consumers to comprehend product information varies for different products. As product reviews are a major source of product-related information, we use product review perception to derive the perceptual schemas. In the paper, we present our three-step method in detail and use it to generate preliminary product clusters. As an exploration of product classification, this research contributes in several perspectives. First, our generated clusters help understand consumer behaviors towards different products. Second, we provide schema prototypes which depict consumers’ perceptual sets towards different products, contributing to both research and practices of online markets. Third, instead of a top-down approach of classifying products, our bottom-up method provides insights of using and mining the value of online textual content.