Ongoing and Past Research Projects
Identifying Key User Perspectives on Mobile Mental Health App Efficacy
Mobile mental health applications are viewed as a promising solution to meet the increasing demand for mental health support. Despite the availability of over ten thousand mental health apps on platforms like Google Play and the Apple App Store, there is limited knowledge about the real-world experiences and concerns of app users. Therefore, we conducted an analysis of user feedback from both the Google Play Store and Apple App Store. Our findings encompass criticisms related to inconsistent content moderation standards and a lack of transparency. Additionally, app-integrated social features and chatbots faced scrutiny for their limited crisis support capabilities. We offer insights for future mental health app developers, emphasizing the need for a comprehensive and centralized app development guideline and exploring the potential of integrating AI technology into mental health chatbots.
Designing Technology for US military Veterans to Facilitate Peer-Mentor Support
US military veterans are a vulnerable population with an elevated risk of mental health issues and suicide. Peer support, especially through mobile technology, has proven effective in addressing mental health-related challenges, but ensuring long-term engagement remains a concern. We are exploring the opportunity of designing technology such as crisis detection, persuasive technology, and peer mentor-mentee matching to enhance engagement in peer support interventions for veterans. We are following community-based participatory research with veterans to identify specific peer support processes that can benefit from this technology and to uncover the underlying community values and needs to guide design.
Designing A Web-based Mobile Mental Health Applications Recommender System
In recent years, the proliferation of mental health apps coincides with the rise of smartphones and digital technologies. However, most of these apps lack proper evaluation and face challenges in reaching users effectively. Users often rely on simplified criteria like app titles, logos, prices, and marketplace ratings to make their choices. To allow consumers to make more informed choices, several systematic frameworks (with and without expert reviews) have been created to rate or rank mental health apps. However, these tools rarely consider the viewpoints of consumers of the MH apps, instead relying on assessments and ratings, which often do not align with those of user needs and concerns. These tools do not allow users to search and comprehend issues faced by previous users of an app, which are provided on online app marketplaces (e.g., Google Play and Apple App Store) through user reviews. On the other hand, online marketplaces do not provide or elaborate expert opinions from healthcare professionals or clinicians. Therefore, there is a need for an all-encompassing guidance framework that includes a visual analysis of both consumer and expert opinions of these apps to understand what matters to consumers and convey these aspects to potential future users, to help them make an informed choice.
Studying How Professional Backgrounds Affect Human-AI Interaction in Algorithmic Crime Mapping
Research into recidivism risk prediction in the criminal legal system has garnered significant attention from HCI, critical algorithm studies, and the emerging field of human-AI decision-making. We focused on algorithmic crime mapping, a prevalent yet underexplored form of algorithmic decision support (ADS) in this context. We interviewed professional crime analysis to understand algorithmic crime mapping practices, used public crime databases to understand commonly mapped crimes, developed visualizations of bias in a potential crime index, examined the use of Lloyd’s algorithm in crime mapping, and conducted a mixed-methods demographically representative study in Milwaukee to understand the use of crime mapping algorithms.
Identifying Digital Privacy Challenges with Shared Mobile Phone Use and Design to Support Digital Privacy with Shared Mobile Phone Use in Bangladesh
Individuals in marginalized communities often share a single device among several users. Yet, the intricate issues surrounding data privacy and conflicts arising from shared device usage have not been studied in depth. So, we investigated privacy practices around the world, including challenges in device sharing, and the benefits, challenges, and trade-offs of a tiered privacy model with a prototype photo gallery app named "Nirapod.” We considered circumstances in which devices are shared, identified social and cultural factors impacting mobile device sharing, designed and tested a prototype photo gallery app, and conducted a three-week deployment in Bangladesh that revealed numerous privacy challenges around device sharing.
Unpacking Digital Privacy Vulnerabilities and Challenges For People in the Global South
As ICT becomes more widespread in the Global South, digital crimes are on the rise. This has prompted governments to implement various surveillance programs. On the other hand, to keep up pace with the rapid growth, citizens sometimes have to seek professional helpers to assist them with different digital tasks. We studied surveillance programs and privacy and security concerns in Bangladesh and explored privacy vulnerabilities through a lens of informal markets in developing nations using ethnographic and qualitative analysis. Through this research, we identified various challenges and insights into the development of digital privacy, safety, and surveillance around the world and explored privacy concepts in the global south, including ownership and identity, and their impact on digital privacy and safety to understand sharing preferences and practices around the world.