The Impact Of Human-Computer Interaction Design On Consumer’s Usage Of Context-Aware Recommender Systems

Technological advancements such as big data have enabled digital devices to gain an ever-increasing amount of contextual information(e.g.,device,location,weather)for recommender systems,producing more personalized and relevant recommendations by adapting them to users’ contextual situations.Those systems that utilize context to provide users with personalized recommendations are known as context-aware recommender systems(CARSs),which have become prevalent in our daily lives.Contrary to conventional recommender systems,CARSs have created a new kind of human-computer interaction(HCI)known as implicit interaction,which is often used to describe interactions that occur without the user explicitly requesting them or being aware of them.The input,processing,and output of HCI in CARSs change significantly as a result of implicit interaction in CARSs,necessitating additional design considerations for this kind of system.CARSs research has historically been heavily technical in nature,and research on this domain from the standpoint of HCI is still in its infancy.Hence,this study explores the impact of HCI design on user decisions from three perspectives:user input,continuous processing,and system output.Study 1 focuses on the impact of user input on users’ usage decisions.There has been little research on how user input influences their usage decisions in CARSs.Some studies have offered technical user input implementations without taking into account the mechanisms by which user input influences usage decisions.Study 1 summarizes various ways in which user input can be employed using user-oriented affordances and extends the application of social interdependence theory from human-human relationships to human-machine interactions.A combination of structural equation modeling analysis and fuzzy set qualitative comparative analysis verifies the importance of user-oriented affordances in CARSs.It is then highlighted that useroriented affordances will promote positive interdependence among users and the system,and thus users’ continued usage of CARSs.Study 2 focuses on the influence of privacy settings for continuous processing on users’ information adoption decisions.Previous studies have mainly focused on the influence of a particular privacy setting,ignoring the interdependence between multiple privacy settings and the impact generated by privacy setting combinations.To this end,Study 2 identified three granular privacy settings as privacy collection settings,privacy usage settings,and privacy protection settings,and decomposed privacy control in CARSs into boundary permeability control,boundary ownership control,and boundary temporary control based upon communication privacy management theory.Results from the analysis of variance identified a cascading effect on the role of privacy settings in privacy control.Additionally,results from the structural equation modeling analysis showed that boundary permeability control influences users’ information adoption decisions via reducing privacy risk probability,whereas boundary ownership control influences consumers’ information adoption decisions via the reduction of both privacy risk probability and privacy risk severity.In addition,no significant effect was found for boundary temporary control on users’ information adoption decisions.Study 3 focuses on the impact of CARSs’ system output features on users’discontinue intentions.Previous research has not systematically summarized the impact of different system output features on user behaviors in CARSs.Grounded on psychological reactance theory,Study 3 proposes that psychological reactance generated by the system output of a CARS may be a significant mechanism for users’intention to discontinue use.Study 3 summarizes the system output features of CARSs through three kinds of context-aware features:context-triggered actions,contextual presentation,and contextual tagging.The results of Study 3 found that three types of context-aware features differ in their impact on different types of perceived threat to freedom.The perceived threat to freedom promotes psychological reactance of users and thus influences their intention to discontinue use.The findings of this study make the following theoretical contributions:(1)this study extends the impact of HCI design on usage decisions in CARSs from the user perspective.While previous studies on CARSs have been explored from a technical perspective,this study examines how user-system interaction affects users’ different usage decisions from three aspects of HCl design,attempting to complement the field of CARSs;(2)this study summarizes the detailed HCI design of CARSs,expanding the dimensions of HCI design;(3)this study enriches mechanisms of how the HCI design of CARSs influences usage decisions;(4)this study extends the application of Social Interdependence Theory and Psychological Reactance Theory;(5)this study extends the empirical research on CARSs.In terms of practical implications,this study firstly reminds designers to focus not only on technical aspects,but also on HCI design.Secondly,this study emphasizes that different types of HCI design have different effects on users’ usage decisions.So CARSs designers can choose to highlight certain types of HCI design according to their purposes.Finally,CARSs designers cannot consider one function in isolation,but also need to consider how different functions affect each other when designing HCI functions.