Authors:
Fanjuan Shi
1
and
Jean-Luc Marini
2
Affiliations:
1
Shanghai Unicore Technology of IOT Co., LTD and Search’XPR SAS, China
;
2
Search’XPR SAS, France
Keyword(s):
Mood Recognition, e-Commerce Recommender System, Behavioral Data Mining, User-centric Systems.
Related
Ontology
Subjects/Areas/Topics:
B2B, B2C and C2C
;
Communication and Software Technologies and Architectures
;
e-Business
;
Enterprise Information Systems
;
Recommendation Systems
;
Society, e-Business and e-Government
;
Software Agents and Internet Computing
;
Web Information Systems and Technologies
Abstract:
This paper presents the result of a controlled experiment studying how mood state can affect the usage of e-commerce recommender system. The authors develop a mood recognition tool to classify online shoppers into stressed or relaxed mood state unobtrusively. By analyzing their reactions to recommended products when surfing on an e-commerce website, the authors make two conclusions. Firstly, stress negatively impacts the usage of recommender system. Secondly, relaxed users are more receptive to recommendations. These findings suggest that mood recognition tool can help recommender systems find the "right time" to intervene. And mood-aware recommender systems can enhance marketer-consumer interaction.