Projects

Human-Robot Interaction: Risk and Trust
User interaction is central to the data analysis process fostered by interactive visual analytic interfaces. However, in many current systems, user interaction is represented as an ephemeral action taken by a user that moves the system from one state to another. User interactions are quantitative bits of the analytic dialog between people, the system, and the data - and when modeled - can be tactfully integrated into visual analytic systems. We propose a library to help researchers and developers capture, interpret, and model interactions in web-based visual analytic tools.
HIV in India is a widely prevalent and complex disease. Treatment is an issue for the economically disadvantaged for financial and social reasons. When designing technology in this context it is important to remember not just the technical requirements but also the effects of stigma and other such contextual factors. NGOs play a key role in the dissemination of awareness and treatment. Thus the outreach workers at the NGO are an important aspect of connecting with the PLHIV. This project is the result of ethnographic research work done over summer '16 and spring '17.
We are looking at identifying instances of Human Trafficking on online marketplaces and review sites. Using textual and visual cues, we are training a classifier to predict the likelihood of being trafficked.
Facilitating the use of smartphone photography to aid psychotherapy
This project aims to study the learning needs of Mandarin Chinese heritage speakers and ultimately create a supplemental tool that will be helpful to them for heritage language learning. The study focuses on any adult of age 18 or older, whose dominant language is English, uses Mandarin Chinese as a heritage language, and is currently located in the United States.
This project hopes to explore ways that solo training can be improved, particularly in beginners. This project involves a full analysis of current methods of solo training and potential deficiencies.
Designing an interface to improve the ability of clinicians to track and understand the mental health status and treatment progress of patients with PTSD
Food journaling refers to the idea of logging food intake with specific details such as calorie count, food ingredients, among other things. Such practice often relies on self-reports, which is prone to recall bias. Even logging during the time of food intake, depending on the context, a user might not be well-informed to log every detail of a food journaling application. Our goal of the project is to improve the user experience of food journaling by minimizing the response burden and utilizing the context of users.
The lack of large-scale, labeled data sets impedes progress in developing robust and generalized predictive models for on-body sensor-based human activity recognition (HAR). Labeled data in human activity recognition is scarce and hard to come by, as sensor data collection is expensive, and the annotation is time-consuming and error-prone. To address this problem, we introduce IMUTube, an automated processing pipeline that integrates existing computer vision and signal processing techniques to convert videos of human activity into virtual streams of IMU data.
Virtual reality as a medium and design space continues to grow, but we suggest that the range of experiences offered rarely stray from the first person perspective. Most experiences, especially video games, have the user directly embody their avatar. In Perspective is a project that explores alternate interfaces that place the user outside of the body of the avatars they inhabit and uses multiple forms of puppetry as design references for new interfaces that blend procedural and physical motion.
For my master's project, I designed In Tune - a tool to improve 1-1, online piano lessons for beginner to intermediate students and their teachers. My design enhances the connection and collaboration between a teacher and a student and helps them stay organized with lesson planning and practice in between lessons. The tool has received positive feedback in evaluation sessions which shows it may enhance how a teacher and student stay organized and connected in this online setting..
As part of the Cognitive Empowerment Program, our team in the Everyday Computing Lab uses in-home technology to learn more about Mild Cognitive Impairment and how it affects adults.
There are many populations who need assistive technologies while driving such as the millions of Americans suffer traumatic brain injuries each year, and the majority of them return to driving at some point following their recovery. However, the residual effects of TBIs can affect perception, cognition, emotion, and motor abilities. In collaboration with the Shepherd Center we are developing software that can help improve the attention and abilities of drivers post-TBI.

Pages