![Sample our Computer Science journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5928/TOC-Subject-Sample-2021-Computer-Science.jpg"})
ABSTRACT
The IoT industry supplies a plethora of Internet connected devices and services supporting smart home automation. However, end-users having little knowledge of the features and possibilities of such technologies, face difficulties in conjuring up useful application scenarios combining such devices and services, thus missing out on potential applications outside those provided by vendors. A remedy for such end-users can potentially be found in crowdsourcing IoT scenario creation. For such an enterprise to be viable it is essential to assess whether crowdsourcing can result in practical and original scenarios. This article reports two studies aiming to establish the practicality and originality of crowdsourced IoT scenarios for smart homes. In the first study, we recruited 102 crowd workers who created 306 scenarios in various categories. We then recruited a second cohort of 620 crowd workers to rate the scenarios’ creativity. In the second study, we evaluated the corpus of IoT scenarios by 20-experienced smart home users recruited through a screening survey. Our results show that the crowd evaluations of originality and creativity are strongly correlated with those of smart home users. Our major IoT-specific findings in relation to creativity are: a) The number of IoT devices and the number of combination of devices impact how creative the scenarios are perceived; b) Workers with self-reported intermediate programming knowledge wrote more creative scenarios when compared to workers having expert knowledge; c) Computational metrics such as text metrics can provide the basis for automated assessment of the scenarios’ creativity. Finally, an inductive thematic analysis of the scenarios revealed interesting themes (e.g., types of rules, automation styles and novel operators) which can serve as a guide for designing more expressive and intuitive end-user development solutions, in the context of IoT.
Notes
8. “A Human Intelligence Task (HIT) is a term introduced on Amazon Mechanical Turk (MTurk) for completing a request of some digital work (e.g., writing a scenario) that can only be performed by people working on MTurk. A HIT represents a single, self-contained task that a Worker can work on, submit an answer, and collect a reward for completing” source: https://www.mturk.com/worker/help.
11. https://atlasti.com/.
12. Fortunately, in our first study, we found that 48.5% of workers had smart home experience though we did not require this from workers for creation task. But in our second study, having a smart home experience was mandatory to evaluate the IoT scenarios; Therefore, we put screening survey in place to hire only experienced workers.
Additional information
Notes on contributors
Tahir Abbas
Tahir Abbas is a PhD candidate at the Industrial Design department of Eindhoven University of Technology, the Netherlands. His research interests include real-time human computation, crowd computing and social robotics. He has a degree in software engineering and is a lecturer, currently on study leave, at Mirpur University of Science & Technology, Pakistan.
Vassilis-Javed Khan
Vassilis-Javed Khan is an assistant professor at the Industrial Design Department of Eindhoven University of Technology in the Netherlands. His current research activities focus on the overlap between crowd computing and design. He has held in the past research positions at Philips Research and Vodafone R&D.
Panos Markopoulos
Panos Markopoulos is a computer scientist specializing in the field of Human Computer Interaction. He is a professor in Design for Behavior Change at the department of Industrial Design in the Eindhoven University of Technology. His current research concerns designing interactive technologies for rehabilitation and for playful learning.