2020 · 12 citations · 11 references
ComputingAutomotive EngineeringComputer SystemsRecommender SystemsTransportationUser Experience DesignContext-aware ComputingContext-aware RecommendationInteraction PatternsMobile Human-computer InteractionRecommendation SystemsUser ModelingContextual Feature RepresentationIn-vehicle User ExperienceInformation OverloadInformation Overload ManagementOverload HandlingUser Context ModelingUser Study EvaluationWorkload
In the last years, the amount of in-vehicle functionalities has drastically increased. Consequently, users frequently have to navigate through a complex menu structure to control a desired functionality. This information overload is typically the cause of distracted driving, which can have unwelcome consequences. The main contribution of this paper is to provide a user behavioral model to represent the recurrent preferences of users for in-vehicle functionalities and to employ it to construct a proactive context-aware recommender system. This work is based on the assumption that most drivers tend to use a given feature within similar contextual environments (i.e. have routines). Therefore, the proposed method aims to reduce the cognitive workload by automatically executing the regularly used features. We believe, this system will not only minimize the number of driver interactions, and hence, the distractions, but also let passengers have a more comfortable and personalized experience. The viability and efficacy of the proposed algorithm are justified by theory and demonstrated on a user study.
11
Pattern recognition and machine learning
Choice Reviews Online · 2007 · 2.7K citations
Recommender system application developments: A survey
Jie Lü, Dianshuang Wu, Mingsong Mao et al. · Decision Support Systems · 2015 · 1.6K citations