(This page will be updated soon)

To tell my research vision briefly, I attempt to say "socio-physical computing".

I envision future computational opportunities which enrich our physical social life, e.g., our social activity and interaction with those right here with us, so that we can see, meet, talk, and touch them in the real world.

We have witnessed that recent advances in mobile social computing contributed a lot to our distant social life, enabling us to be more socially interactive with distant people. Now I ask, "what can we do for our physical social life, which has been undoubtedly natural and prevalent since the beginning of humankind?"

Pursuing my research vision toward socio-physical computing, I have designed and deployed a number of multi-personal sensing systems for various domains, e.g., clinical, educational, and entertainment domains, as well as developed novel mobile sensing platforms for socially interacting users. It is always fascinating for me to explore different domains and pave brand-new roads connecting those with computational foundations.

Interpersonal Touch towards Vibrant Social Life Everyday Clinical Care through Interaction Sensing Behavioral Relativity in Educational Social Groups Pervasive Social Exergames
Mobile Sensing Platforms for Multi-personal Settings


Interpersonal Touch towards Vibrant Social Life


Interpersonal touch is our most primitive social language strongly governing our emotional well-being. Despite the positive implications of touch in many facets of our daily social interactions, we find wide-spread caution and taboo limiting touch-based interactions in workplace relationships that constitute a significant part of our daily social life. In this first work, we explore new opportunities for ubicomp technology to promote a new meme of casual and cheerful interpersonal touch such as high-fives towards facilitating vibrant workplace culture. Specifically, we propose High5, a mobile service with a smartwatch-style system to promote high-fives in everyday workplace interactions. We first present initial user motivation from semi-structured interviews regarding the potentially controversial idea of High5. We then present our smartwatch-style prototype to detect high-fives based on sensing electric skin potential levels. We demonstrate its key technical observation and performance evaluation.




Related publications:


Everyday Clinical Care through Interaction Sensing


This is my new research project to expand mobile sensing toward reinforcing the self-effort of the patients and their close social peers in everyday life. For the patients visiting the clinicians in a regular basis for a long-term treatment, the formative clinical treatments are undoubtedly essential but highly limited in terms of the temporal and spatial coverage. Except a few hours a week in the clinical facilities, the patients are on their own. We focus on the large clinical potential within such unattended moments in everyday life. It is a general advice that, when the patients themselves’ everyday efforts are appended to existing formative treatments or medication, better clinical outcomes can be expected. We envision new mobile- and social-sensing services to provide in-situ contextualized interventions overlaid in the patients' everyday life situations, effectively promoting clinical favorable behaviors and life-styles and eventually expediting the treatment procedure.

As the first work, we present TalkBetter, a mobile service to help the family of young patients with language delay. Language delay is a developmental problem of young children who do not acquire language ability as expected for their chronological ages. Without timely intervention, language delay can act as a lifelong risk factor, which includes a learning disorder, a social skill deficit, and a low socioeconomic status. Based on extensive field study with speech-language pathologists, we explored new mobile social computing opportunities to reinforce everyday effort for the remediation of language delay. Through mobile conversation sensing and in-situ intervention, we try to expedite therapeutic implications in everyday family conversation with the child with language delay.




Related publications:


Behavioral Relativity in Educational Social Groups


This research explores new opportunities for ubiquitous technologies to contribute to childhood education environment like kindergartens. We envision that kindergartens are fascinating observational settings as children of similar ages share their environment and experiences, practicing social interactions and receiving preliminary education together. Our intuition is that focusing on how each child behaves or responds distinctly under shared situations, namely inter-child behavioral relativity, would lead to early and rich understanding of each child’s character, talents, or needs.

To this end, wearable sensor technology can be well-fit as it delivers quantitative bases reflecting each child’s physical behaviors, which is an essential prerequisite to make comparisons across them and evaluate their relativity. Our particular choice of their behavioral aspects as the major means of understanding comes from the developmental nature of children that behavioral expressions still dominate their total expressiveness. Diving into the real kindergarten classes and field activities, we have been developing highly specialized application and framework system monitoring situation-specific inter-child behavioral relativity.

Related publications:


Pervasive Social Exergames


This research explores the potential of pervasive social exertion games transforming solitary exercises into socially enjoyable activities. It is difficult for many people with their busy and erratic urban lifestyles to engage in such social sports frequently as such activities take place at a strict designated time and place. As a result, many people perform physical training in the form of a solitary exercise that can be done at a more convenient time and place, such as treadmill running at home. These solitary exercises, however, are mostly monotonous, repetitive and boring, and, moreover, performed in isolation. Such disadvantages discourage people from exercising regularly, depriving them of the many known benefits of. Even worse, solitary exercise does not satisfy the need for social interaction, which is important for an individual’s health and sense of well-being.
We have developed a number of social exergame systems for many different aerobic exercises - running, cycling, swimming, jump-roping, etc. We also developed ExerLink, a cross-exercise gaming platform to support multiple players each with a different favorite exercise can enjoy well-balanced collaborative exergame experience simultaneously.



Related publications:

Mobile Sensing Platforms for Multi-personal Settings


Powered by the prevalence of personal mobile devices such as smartphones, mobile sensing applications are increasingly emerging. Today, mobile services like Google Now continuously sense our location and activity, delivering proactive services based on our context without requiring our interaction. However, many of these applications are yet in experimental stages - leveraging limited contextual information and requiring significant computational and energy resources. 
We explore the breakthrough for future mobile sensing from a novel perspective - leveraging the highly social nature of humans. Meeting people, making interactions, and doing activities together comprise significant part of our everyday routine. As our initial explorations to date, we present CoMon and SocioPhone.
In CoMon, we present a cooperative ambience sensing platform to achieve mutual resource benefits for multiple users who are sharing co-located social activities. We built the context sharing infrastructure across sensing applications of different users, the cost-benefit analysis model, the access control, and so on.

In SocioPhone, we envision a new class of future applications, namely interaction-aware application. To enable and support those applications, we present a novel face-to-face interaction monitoring platform, and propose the meta-linguistic conversational turn-takings as the key contextual element. To realize it in everyday settings with today's commodity mobile devices, we built highly agile and light-weight meta-linguistic conversation monitoring engine.

Subscribe to: Posts (Atom)