Intention-Based Online Consumer Classification for Recommendation and Personalization. Hot Topics in Web Systems and Technologies

Fanjuan Shi (1) , Chirine Ghedira (2)

1 Laboratoire de Recherche Magellan
2 SOC - Service Oriented Computing

Consumers' online shopping behaviors are mostly determined by their intentions. Thus, the knowledge of consumer intention can help online marketers to enhance sales conversion rate and reduce ineffective marketing communications. Current personalization and recommendation techniques do not pay enough attention to various consumer intentions. The taxonomy of online shopping intention and the method to predict intention in real time are yet to be developed. Based on unsupervised and supervised learning techniques, this paper proposes an intention prediction model to fulfill the research gap. Empirical results suggest that the proposed model is able to classify intentions precisely. Accordingly, we discuss the implication and provide some managerial suggestions to online marketers who seek to implement some intention-based personalization methods.

Mots clés: Internet; consumer behaviour; unsupervised learning; intention-based online consumer classification; online marketers; online shopping behaviors; supervised learning techniques; unsupervised learning techniques; Classification algorithms; Machine learning algorithms; Prediction algorithms; Real-time systems; Taxonomy; Web pages; classification; clustering; consumer intention; data mining; e-commerce; machine learning

Domaines: Informatique [cs]

Fanjuan Shi, Chirine Ghedira. Intention-Based Online Consumer Classification for Recommendation and Personalization. Hot Topics in Web Systems and Technologies. HOTWEB, Oct 2016, Washington, DC, United States. pp.36-41, ⟨10.1109/HotWeb.2016.15⟩. ⟨hal-01477372⟩