Supporting User Understanding and Engagement in Designing Intelligent Systems for the Home.

Designing Intelligent Systems for the Home;Information and Library Science;Social Sciences;Information

Yang, RayoungAckerman, Mark Steven ;

University of Michigan

关键词: Designing Intelligent Systems for the Home;    Information and Library Science;    Social Sciences;    Information;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/133318/rayang_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

With advances in computing, networking and sensing technology, our everyday objects have become more automated, connected, and intelligent. This dissertation aims to inform the design and implementation of future intelligent systems and devices. To do so, this dissertation presents three studies that investigated user interaction with and experience of intelligent systems. In particular, we look at intelligent technologies that employ sensing technology and machine learning algorithm to perceive and respond to user behavior, and that support energy savings in the home. We first investigated how people understand and use an intelligent thermostat in their everyday homes to identify problems and challenges that users encounter. Subsequently, we examined the opportunities and challenges for intelligent systems that aimed to save energy, by comparing how people’s interaction changed between conventional and smart thermostats as well as how interaction with smart thermostats changed over time. These two qualitative studies led us to the third study. In the final study, we evaluated a smart thermostat that offered a new approach to the management of thermostat schedule in a field deployment, exploring effective ways to define roles for intelligent systems and their users in achieving their mutual goals of energy savings. Based on findings from these studies, this dissertation argues that supporting user understanding and user control of intelligent systems for the home is critical allowing users to intervene effectively when the system does not work as desired. In addition, sustaining user engagement with the system over time is essential for the system to obtain necessary user input and feedback that help improve the system performance and achieve user goals. Informed by findings and insights from the studies, we identify design challenges and strategies in designing end-user interaction with intelligent technologies for the home: making system behaviors intuitive and intelligible; maintaining long-term, easy user engagement over time; and balancing interplay between user control and system autonomy to better achieve their mutual goals.