Kate is a PhD student in the Paul G. Allen School of Computer Science and Engineering at the University of Washington. She is advised by Professor Jennifer Mankoff. She completed her undergraduate studies at USC, where she double-majored in Computer Science and Business Administration, as well as received her master’s degree in Computer Science. She is an NSF CSGrad4US fellow.
She is interested in studying the intersection of digital and physical technologies that empower those with disabilities or illnesses. Her recent research focuses on generative AI and accessibility, seeking to gain a deeper understanding of the opportunities for improving access as well as identifying areas for improvement.
Hello! I am a Masters student in the school of Electrical & Computer Engineering.
I am broadly interested in Technology, the world and philosophy. Although my specific research interests are still maturing, HCI and AI are the fields that captivate me currently.
Working at the Intersection of Race, Disability, and Accessibility
Examinations of intersectionality and identity dimensions in accessibility research have primarily considered disability separately from a person’s race and ethnicity. Accessibility work often does not include considerations of race as a construct, or treats race as a shallow demographic variable, if race is mentioned at all. The lack of attention to race as a construct in accessibility research presents an oversight in our field, often systematically eliminating whole areas of need and vital perspectives from the work we do. Further, there has been little focus on the intersection of race and disability within accessibility research, and the relevance of their interplay. When research in race or disability does not mention the other, this work overlooks the potential to better understand the full nuance of marginalized and “otherized” groups. To address this gap, we present a series of case studies exploring the potential for research that lies at the intersection of race and disability. We provide examples of how to integrate racial equity perspectives into accessibility research, through positive examples found in these case studies and reflect on teaching at the intersection of race, disability, and technology. This paper highlights the value of considering how constructs of race and disability work alongside each other within accessibility research studies, designs of socio-technical systems, and education. Our analysis provides recommendations towards establishing this research direction.
Dashboards are frequently used to monitor and share data across a breadth of domains including business, finance, sports, public policy, and healthcare, just to name a few. The combination of different components (e.g., key performance indicators, charts, filtering widgets) and the interactivity between components makes dashboards powerful interfaces for data monitoring and analysis. However, these very characteristics also often make dashboards inaccessible to blind and low vision (BLV) users. Through a co-design study with two screen reader users, we investigate challenges faced by BLV users and identify design goals to support effective screen reader-based interactions with dashboards. Operationalizing the findings from the co-design process, we present a prototype system, Azimuth, that generates dashboards optimized for screen reader-based navigation along with complementary descriptions to support dashboard comprehension and interaction. Based on a follow-up study with five BLV participants, we showcase how our generated dashboards support BLV users and enable them to perform both targeted and open-ended analysis. Reflecting on our design process and study feedback, we discuss opportunities for future work on supporting interactive data analysis, understanding dashboard accessibility at scale, and investigating alternative devices and modalities for designing accessible visualization dashboards.
Arjun Srinivasan, Tim Harshbarger, Darrell Hilliker and Jennifer Mankoff: University of Washington (2023): “Azimuth: Designing Accessible Dashboards for Screen Reader Users” ASSETS 2023.
Ben is an incoming second-year undergraduate at the University of Washington, majoring in computer science. He has prior programming experience in mobile development, machine learning, and data visualization. He is excited to learn more about data science and how it can be used to inform decisions for everyday life. In the near future, he also hopes to dive into computer vision and databases. His outside interests include playing and watching basketball, listening to music, and running. He will be working on analyzing the associations between various student groups’ behavior, academics, and well-being.
I completed my Bachelor’s degree in Engineering from Netaji Subhas Institue of Technology, New Delhi. I majored in Mechanical Engineering; however, I found my true calling in working with software. Currently, I am a master’s student at the University of Washington majoring in Information Management.
Many individuals with disabilities and/or chronic conditions experience symptoms that may require intermittent or on-going medical care. However, healthcare is often overlooked as an area where accessibility needs to be addressed to improve physical and digital interactions between patients and healthcare providers. We discuss the challenges faced by individuals with disabilities and chronic conditions in accessing physical therapy and how technology can help improve access. We interviewed 15 people and found both social (e.g. financial constraints, lack of accessible transportation) and physiological (e.g. chronic pain) barriers to accessing physical therapy. Our study suggests that technology interventions that are adaptable, support movement tracking, and community building may support access to physical therapy. Rethinking access to physical therapy for people with disabilities or chronic conditions from a lens that includes social and physiological barriers presents opportunities to integrate accessibility and adaptability into physical therapy technology.
Speechreading is the art of using visual and contextual cues in the environment to support listening. Often used by d/Deaf and Hard-of-Hearing (d/DHH) individuals, it highlights nuances of rich communication. However, lived experiences of speechreaders are underdocumented in the literature, and the impact of online environment and interaction of captioning with speechreading has not been explored. To bridge these gaps, we conducted a three-part study consisting of formative interviews, design probes and design sessions with 12 d/DHH individuals who speechread.
We present an ethnographic study of a maker community that conducted safety-driven medical making to deliver over 80,000 devices for use at medical facilities in response to the COVID-19 pandemic. To achieve this, the community had to balance their clinical value of safety with the maker value of broadened participation in design and production. We analyse their struggles and achievement through the artifacts they produced and the labors of key facilitators between diverse community members. Based on this analysis we provide insights into how medical maker communities, which are necessarily risk-averse and safety-oriented, can still support makers’ grassroots efforts to care for their communities. Based on these findings, we recommend that design tools enable adaptation to a wider set of domains, rather than exclusively presenting information relevant to manufacturing. Further, we call for future work on the portability of designs across different types of printers which could enable broader participation in future maker efforts at this scale.
There is a growing body of research revealing that longitudinal passive sensing data from smartphones and wearable devices can capture daily behavior signals for human behavior modeling, such as depression detection. Most prior studies build and evaluate machine learning models using data collected from a single population. However, to ensure that a behavior model can work for a larger group of users, its generalizability needs to be verified on multiple datasets from different populations. We present the first work evaluating cross-dataset generalizability of longitudinal behavior models, using depression detection as an application. We collect multiple longitudinal passive mobile sensing datasets with over 500 users from two institutes over a two-year span, leading to four institute-year datasets. Using the datasets, we closely re-implement and evaluated nine prior depression detection algorithms. Our experiment reveals the lack of model generalizability of these methods. We also implement eight recently popular domain generalization algorithms from the machine learning community. Our results indicate that these methods also do not generalize well on our datasets, with barely any advantage over the naive baseline of guessing the majority. We then present two new algorithms with better generalizability. Our new algorithm, Reorder, significantly and consistently outperforms existing methods on most cross-dataset generalization setups. However, the overall advantage is incremental and still has great room for improvement. Our analysis reveals that the individual differences (both within and between populations) may play the most important role in the cross-dataset generalization challenge. Finally, we provide an open-source benchmark platform GLOBEM – short for Generalization of LOngitudinal BEhavior Modeling – to consolidate all 19 algorithms. GLOBEM can support researchers in using, developing, and evaluating different longitudinal behavior modeling methods. We call for researchers’ attention to model generalizability evaluation for future longitudinal human behavior modeling studies.
, 
