Abstracts:

Matthew Hong: Personalizing Health Management Through Human-Centered Data Augmentation

During complex chronic treatment, adolescent patients (ages 10-19) must communicate all illness needs to the care team so they can access relevant health resources when most needed. This communication is challenging because patients, family caregivers and clinicians have unmatched experiences, conceptions and linguistic representations of indicators of health. Most importantly, patients lack the means to capture and represent their felt illness experience. My colleagues and I addressed these challenges by advancing personalized computing technology and human-centric methods that inform collaborative approaches for managing personal health data. In this talk, I will describe how technology can be designed to effectively scaffold patients’ gradual participation in managing their illness. I draw from Health Informatics, Participatory Design, and Human-Computer Interaction to show how we can augment clinically-generated, and patient-generated data in ways that cater to personal health needs. I will discuss how human-centered data-augmentation can help designers create intelligent systems to improve chronic care for pediatric patients.

Lara Martin: Understanding the Technological and Experiential Requirements of Improvisational Storytelling Agents

Although we are currently riding a technological wave of personal assistants, many of these agents still struggle to communicate appropriately. Humans are natural storytellers, so it would be fitting if artificial intelligence could tell stories as well. Automated story generation is an area of AI research that aims to create agents that tell “good” stories. Previous story generation systems use planning to create new stories, but these systems require a vast amount of knowledge engineering. The stories created by these systems are coherent, but only a finite set of stories can be generated. In contrast, very large language models have recently made the headlines in the natural language processing community. Though impressive on the surface, these models begin to lose coherence over time. My research looks at various techniques of automated story generation, focusing on the perceived creativity of the generated stories. Here, I define a creative product as one that is both novel and useful. In my dissertation, I theorize that a jointly probabilistic and causal model will provide more creative stories for readers of stories generated from an improvisational storytelling system than solely probabilistic or causal models.

Emily Wall: Mitigating Implicit Human Bias in Visual Analytics

Implicit bias is a term used to describe the way that our culture, experiences, and stereotypes can unconsciously impact our attitudes and decision making. Such biases, like racial or gender bias, can impact decision making in critical ways, propagating long-standing institutional and systemic biases. However, as decision making is increasingly taking place with the aid of data-driven visual representations (including interactive visualization tools like Tableau, among others), we are afforded a new opportunity with respect to the detection and mitigation of implicit biases. In this talk, I describe (1) how user interactions with data can be used to approximate implicit biases and (2) how visualization systems can be designed to make implicit biases more explicit by increasing awareness. By creating systems that promote real-time awareness of bias, people can reflect on their behavior and decision making and ultimately engage in a less-biased decision making process.