Exploring the Design Space of User-System Communication for Smart-home Routine Assistants

Chiang, Yi-Shyuan; Chang, Ruei-Che; Chuang, Y. T.; Chou, Shih-Ya; Lee, Hao-Ping; Lin, I-Ju; Chen, Jian-Hua Jiang; Chang, Yi‐Jun

doi:10.1145/3313831.3376501

Cited by 18 publications

(2 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Instead of making decisions fully on the user's behalf, a context-aware system should communicate and collaborate with users. Recent work has thus attempted to provide useful information about the system to users [26,30,37] and designed interfaces that establish constructive collaboration between systems and users [8,20,39].…”

Section: Context-aware Computingmentioning

confidence: 99%

Trace2TAP

Zhang

Morkved

et al. 2020

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.

View full text Add to dashboard Cite

Two common approaches for automating IoT smart spaces are having users write rules using trigger-action programming (TAP) or training machine learning models based on observed actions. In this paper, we unite these approaches. We introduce and evaluate Trace2TAP, a novel method for automatically synthesizing TAP rules from traces (time-stamped logs of sensor readings and manual actuations of devices). We present a novel algorithm that uses symbolic reasoning and SAT-solving to synthesize TAP rules from traces. Compared to prior approaches, our algorithm synthesizes generalizable rules more comprehensively and fully handles nuances like out-of-order events. Trace2TAP also iteratively proposes modified TAP rules when users manually revert automations. We implemented our approach on Samsung SmartThings. Through formative deployments in ten offices, we developed a clustering/ranking system and visualization interface to intelligibly present the synthesized rules to users. We evaluated Trace2TAP through a field study in seven additional offices. Participants frequently selected rules ranked highly by our clustering/ranking system. Participants varied in their automation priorities, and they sometimes chose rules that would seem less desirable by traditional metrics like precision and recall. Trace2TAP supports these differing priorities by comprehensively synthesizing TAP rules and bringing humans into the loop during automation.

show abstract

Section: Context-aware Computingmentioning

confidence: 99%

Trace2TAP

Zhang

Morkved

et al. 2020

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.

View full text Add to dashboard Cite

show abstract

“…First, the home product industry consists of a market of over US$400bn in the USA alone (Statista, 2021). Second, home products are designed for household routines, such as vacuuming and cooking (Chiang et al , 2020), and VAF can potentially automate part of household chores if not all (Pan, 2022; Thormundsson, 2022). More importantly, consumers have established schema for routine tasks, so identifying opportunities and barriers in using voice-assisted home products is critical for developing a human-centric approach as voice assistant technology evolves.…”

Section: Introductionmentioning

confidence: 99%

Why can’t you be all talk to get things done? Consumer acceptance of voice-assisted products

Xiao,

Tan

2023

JPBM

View full text Add to dashboard Cite

Purpose Voice assistant technology represents one of the most radical artificial intelligence innovations. Drawing on the processing fluency theory and consumer learning literature, this study aims to explore how consumer acceptance of new products is influenced by voice assistant function (VAF), along with the impacts of role clarity and learning modality. Design/methodology/approach Four between-subjects experimental studies were conducted. Study 1 tested the main effect of VAF on consumer acceptance. Study 2 included role clarity as a mediator between VAF and consumer acceptance. Study 3 examined the moderation effect of learning modality and contrasted the effectiveness of experiential and verbal learning in helping increase consumer acceptance. Study 4, as a post hoc study, tested serial mediations to validate whether processing fluency was indeed the mechanism explaining the indirect relationship between VAF and consumer acceptance via role clarity. Findings The negative impact of VAF on consumer acceptance was demonstrated in all four studies. Studies 2 and 3 showed VAF decreased role clarity which further influenced consumer acceptance. Moreover, Study 3 evidenced that experiential learning was more effective than verbal learning in increasing consumer acceptance of voice-assisted products via role clarity. Study 4 demonstrated that VAF decreased role clarity, which in turn decreased processing fluency, leading to lower consumer acceptance. Originality/value This research views the usage of voice-assisted products as a coproduction process between consumers and the VAF. Accordingly, findings provide novel insights into processing fluency of tasks assisted by VAF through the lens of role clarity and learning modality, which enriches the understanding of potential barriers and opportunities for consumers to accept voice-assisted products.

show abstract