Authors:
Yi-Cheng Ku
1
;
Chih-Hung Chan
1
and
Chin-Sheng Yang
2
Affiliations:
1
Providence University, Taiwan
;
2
College of Informatics and Yuan Ze University, Taiwan
Keyword(s):
Recommender Systems, Product Type, Self-monitoring, Purchase Intention, Laboratory Experiment.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Electronic Commerce
;
Enterprise Information Systems
;
Internet Systems Performance
;
Knowledge Discovery and Information Retrieval
;
Knowledge-Based Systems
;
Software Agents and Internet Computing
;
Symbolic Systems
;
User Profiling and Recommender Systems
Abstract:
Many e-stores offer product recommendation service to increase sales and customers’ satisfaction. However, the performance of recommendation system will be influenced by the consumers’ judgments on recommended products. The purpose of this study is to investigate the effect of product type and recommendation approach on consumers’ intention to purchase recommended products. A laboratory experiment was conducted to collect empirical data and ANCOVA was adopted to test the research hypotheses. We found that there are significant interaction effects between recommendation approach and recommended product type on consumers’ willingness to buy the recommended products and their disconfirmation of the recommended products, while the consumers’ self-monitoring degree was a covariate variable. Before experiencing the products, consumers’ willingness to buy search goods recommended by the top-N approach is significantly higher than those recommended by the collaborative-filtering recommendati
on approach. On the other hand, after experiencing the products, consumers’ disconfirmation of the experience goods recommended by the top-N approach is also significantly higher than those recommended by the collaborative-filtering recommendation approach. Besides, among the products recommended by the top-N approach, the disconfirmation of experience goods is significantly higher than that of search goods. The results of this study provide valuable implications for researchers and practitioners.
(More)