Most existing approaches in Context-Aware Recommender Systems (CRS) focus on recommending relevant items to users taking into account contextual information, such as time, location, or social aspects. However, few of them have considered the problem of user's content dynamicity. We introduce in this paper an algorithm that tackles the user's content dynamicity by modeling the CRS as a contextual bandit algorithm and by including a situation clustering algorithm to improve the precision of the CRS. Within a deliberately designed offline simulation framework, we conduct evaluations with real online event log data. The experimental results and detailed analysis reveal several important discoveries in context aware recommender system.
Abstract. The wide development of mobile applications provides a considerable amount of data of all types. In this sense, Mobile Context-aware Recommender Systems (MCRS) suggest the user suitable information depending on her/his situation and interests. Our work consists in applying machine learning techniques and reasoning process in order to adapt dynamically the MCRS to the evolution of the user's interest. To achieve this goal, we propose to combine bandit algorithm and case-based reasoning in order to define a contextual recommendation process based on different context dimensions (social, temporal and location). This paper describes our ongoing work on the implementation of a MCRS based on a hybrid-ε-greedy algorithm. It also presents preliminary results by comparing the hybrid-ε-greedy and the standard ε-greedy algorithm.Keywords: Machine learning, contextual bandit, personalization, recommender systems, exploration/exploitation dilemma. IntroductionMobile technologies have made access to a huge collection of information, anywhere and anytime. Thereby, information is customized according to users' needs and preferences. This brings big challenges for the Recommender System field. Indeed, technical features of mobile devices yield to navigation practices which are more difficult than the traditional navigation task. A considerable amount of research has been done in recommending relevant information for mobile users. Earlier techniques [8,10] are based solely on the computational behavior of the user to model his interests regardless of his surrounding environment (location, time, near people). The main limitation of such approaches is that they do not take into account the dynamicity of the user's context. This gives rise to another category of recommendation techniques that tackle this limitation by building situation-aware user profiles. However, these techniques have some problems, namely how to recommend information to the user in order to follow the evolution of his interest.In order to give Mobile Context-aware Recommender Systems (MCRS) the capability to provide the mobile user information matching his/her situation and adapted to the evolution of his/her interests, our contribution consists of mixing bandit algorithm (BA) and case-based reasoning (CBR) methods in order to tackle these two issues:
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.