Daye Kang, Zhuolun Han, Jiahe Tian, Muhan Zhang, and Jeff Rzeszotarski 

This paper explores the potential role of AI, e.g., large language models (LLMs), in supporting theme development in thematic analysis. While prior applications of AI in qualitative data analysis have focused on supporting coding, we investigate whether LLMs can effectively contribute as collaborators in the more abstract and conceptual phases of qualitative analysis, specifically theme development. Despite growing interest in AI as a collaborator in theme development, there is limited empirical evidence on designing AI-assisted tools while supporting user autonomy and understanding researcher interaction with AI-assisted theme development. To address this gap, we designed ThemeViz, an interactive system that uses GPT-4 to generate and visualize multiple versions of themes based on user input while allowing researchers to maintain control through manual coding and theme development. Our study examines the effectiveness of this human-AI collaboration approach in iterative theme development and its implications for future designs.

[PDF] [DOI]

Daye Kang, Jingjin Li, Gilly Leshed, Jeff Rzeszotarski and Xi Lu

Polycystic ovary syndrome (PCOS) is a common hormonal disorder affecting 11-13% of women of reproductive age, characterized by a wide range of symptoms (e.g., menstrual irregularity, acne, and obesity) that varies among individuals. While self-tracking tools help PCOS patients to monitor their symptoms and find personalized treatment, they often focus on regular periods of healthy women with inadequate support for the 1) personalization and 2) long-term holistic tracking necessary for managing complex chronic conditions like PCOS. To bridge this gap, the first author (who has PCOS) conducted an autoethnographic study of holistic self-tracking over a period of ten months in an effort to manage her condition. Our results highlight the challenges of personalized, holistic, long-term tracking in medical, socio-cultural, temporal, technical, and spatial contexts. Based on these insights, we provide design implications for tracking tools that are more inclusive and sustainable.

[PDF] [DOI]

Daye Kang, and Jeff Rzeszotarski

Qualitative UX research practitioners analyze data to comprehend users' needs and synthesize implications for future design. Working with multiple stakeholders is inevitable in modern product development, and in industry, coordination and collaboration add pressures to already laborious qualitative data analysis tasks. In our work, we investigate how multi-stakeholder collaboration specifically affects industry qualitative data analysis practices. We investigate the challenges practitioners face, limitations of current tools, and opportunities for computer-supported cooperative work. 

[PDF] | [DOI] 

Daye Kang, Tony Ho, Nicolai Marquardt, Bilge Mutlu, Andrea Bianchi​

ToonNote is a novel technique for representing computational notebooks in the form of interactive data comics. ToonNote provides a high-level, curated narrative of the dataset in Comic View. Readers can focus on data storytelling, and not be hindered by code, unnecessary outputs, or markdown and can switch back to the Notebook View when needed. 

PDF | GitHub | DOI
 

Yoonji Kim, Youngkyung Choi, Daye Kang, Minkyeong Lee, Tek-Jin Nam, Andrea Bianchi

Physical computing is a complex activity that consists of different but tightly coupled tasks: programming and assembling hardware for circuits. We propose a general-purpose prototyping tool based on conversation. HeyTeddy guides users during hardware assembly by providing additional information on requests or by interactively presenting the assembly steps to build a circuit. 

VIDEO | PDF | DOI 

Yujin Lee, Jee Bin Yim, Daye Kang, Hyeon-Beom Yi, Daniel Saakes

In this research, we explored the effects of the internal structure on the texture of chocolate by designing, building, and testing chocolates with different internal structures and internal chocolate percentage. Multiple layers of patterned chocolate were stacked as a fabrication method. Each layer was fabricated using a silicone mold that was made using a 3D printed model. We created seven types of chocolate variations through this system.

​

PDF | DOI 

Daye Kang, Hye-Ryeong Kim, Ji-Hae Lee, Jae Young Yun.

Treating dyslexia during childhood is very important. In Korea, this is not the case since the condition is not well understood. Also, there is a limited number of medical centers that offer treatment for it. We propose an education app for dyslexic children ages 5-10 in Korea that can help these children through active learning. By teaching the children Korean phonics using 'nonwords' in a fun and engaging way, they can be treated at home instead of visiting the medical centers.

​

Daye Kang, Hye-Min Choi, Ka-Hyun Kim, Younjoon Lee.

Hospitalized children are often asked to regularly walk in the hallway to stay active. However, the hallway is usually dull and empty. We created wallpapers and floor footprint stickers that use nudge design to be fun and engaging. Footprint stickers promote walking and leg stretches while the wallpaper contains stories that can add fun moments. Children are guided through stretching activities indicated on the wall.