Ceara Byrne: Technology for Working and Service Dogs

In the Animal Computer Interaction (ACI) lab, I create and study technologies that improve communication between working dogs, such as dogs trained for search and rescue, and their handlers. In particular, my research focuses on improving the outcomes of service and working dog training. Not all dogs that go into these training programs as puppies have the temperament to become successful assistance and working animals. However, it is very difficult to determine if a dog has a temperament suitable for a service or working animal early on in life. That is where my research comes in. In my work, I investigate how aspects of canine temperament can be detected from interactions with sensors, often placed inside of dog toys that I design and build. After running tests where dogs interact with these sensors, I develop models that use sensor data to predict the success of assistance dogs in advanced training.

Azra Ismail: Human-Centered Design of Artificial Intelligence Systems for Frontline Health

There has been growing interest in the application of Artificial Intelligence (AI) in frontline health, motivated by a shortage of skilled medical experts and medical equipment, particularly in the Global South. The global COVID-19 pandemic has drawn attention to the potential for these efforts, but also their many limitations. These systems may increase the work burden on frontline health workers, many of whom are women engaged in underpaid and invisible care and data work. In this talk, I will examine the AI for Global Health discourse, the gaps in current efforts, and opportunities for design, while centering the perspectives of frontline health workers. I will draw on data from three years of ethnographic fieldwork that I have conducted with women frontline health workers and women from underserved communities in Delhi (India), as well as an extensive literature review of ongoing AI efforts in this space. Finally, I will draw on a rich body of literature on Human-Computer Interaction for Development (HCI4D), post-development critique, and transnational feminist theory to discuss lessons for AI efforts that target social good more broadly.

Koustuv Saha: Computational and Causal Approaches on Social Media and Multimodal Sensing Data: Examining Wellbeing in Situated Contexts

A core aspect of our social lives is often embedded in situated communities, such as our workplaces, neighborhoods, localities, and school/college campuses. The inter-connectedness and inter-dependencies of our interactions, experiences, and concerns intertwine our situated context with our wellbeing. A better understanding of our wellbeing and psychosocial dynamics will help us devise strategies to address our wellbeing through proactive and tailored support strategies. However, existing methodologies to assess wellbeing suffer from limitations of scale and timeliness. Parallelly, given its ubiquity and widespread use, social media can be considered a “passive sensor” that can act as a complementary source of unobtrusive, real-time, and naturalistic data to infer wellbeing. This talk will present an overview of computational and causal approaches for leveraging social media in concert with complementary multimodal sensing data to examine wellbeing in situated contexts. This talk will show how theory-driven computational methods can be applied on unique social media and complementary multimodal sensing data to capture attributes of human behavior and psychosocial dynamics in situated communities.