Skip to Main Content

Dissertation Defenses

Our Ph.D. students work hard for years honing their research. This effort culminates in an oral dissertation where they present, explain, and defend their ideas. The final defense of the dissertation is conducted by an examination committee recommended by the graduate faculty advisor and approved by the dean of The Graduate School. All final oral examinations are open to members of the graduate faculty.

Below is the published list of upcoming or recent dissertation defenses for Information Systems students.

Upcoming Defenses

Title: A Self-supervised Longitudinal Activity and Cognitive Health Monitoring Framework for Older Adults.
Student: Sreenivasan Ramasamy Ramamurthy
Date/Time: Friday, May 14, 2021, 10:00 AM – 12:00 PM
Link: https://umbc.webex.com/umbc/j.php?MTID=mfb14844740ad097823286592fb1a2aee

Meeting number:120 459 9123

Password:Xts4KfTP

Committee:
Dr. Nirmalya Roy (Chair), UMBC
Dr. Aryya Gangopadhyay, UMBC
Dr. Zhiyuan Chen, UMBC
Dr. Elizabeth Galik, UMB

Dr. Bivas Mitra, IIT Kharagpur

Abstract

The design, development, and deployment of human activity recognition (HAR) and personalized health monitoring applications in smart home environments have been accelerated with the recent penetration of wearable and off-the-shelf consumer-grade Internet-of-Things (IoT) devices, along with the advancement of artificial intelligence (AI) and machine learning (ML) techniques. In recent years, substantial research effort has been conducted to improve human activity recognition and physiological health monitoring using various deep learning paradigms (e.g. supervised, unsupervised, semi/self-supervised learning algorithms) and bio-signals (e.g. photoplethysmograph, electrocardiogram, galvanic skin response, electrodermal activity, etc). However, early detection of functional and behavioral health deteriorations due to the cognitive health decline using only the accelerometry data has been lightly investigated. In this thesis, we propose to study the relationships between activities and cognitive health declines for older adults longitudinally using accelerometry signals while focusing on addressing the challenges arising from micro- and macro- activity sensing, user-agnostic activity model learning, and personalized health monitoring.

To support this effort, we performed an extensive data collection drive to gather sensor-based activity data and survey-based clinical assessment on functional, behavioral, and cognitive health from 25 older adults residing in their own homes in a retirement community center. This collection of real data using our in-house smart home sensing systems from community-dwelling older adults helped us to realize the challenges associated with the missed detection of specific activities during the sensing phase, appropriate demarcation of activity transition points, availability of abundant unlabeled data, and severe paucity of labeled data. To address the above challenges, particularly the availability of abundance unlabeled data in a practical setting, first, we propose a self-taught learning approach that helps to increase the precision and reduce the missed detections in activity recognition and sensing, respectively. To further improve the activity recognition performance while addressing the demarcation of the activity transition points across multiple older adults, we propose a user-invariant scalable activity recognition framework relying on contrastive learning. Next, we propose a multi-task learning framework that helps exploit the correlations between the fine-grained functional and cognitive health decline measurement metrics.

Our proposed self-supervised activity and cognitive health monitoring framework relying on self-taught, contrastive, and multi-task learning approaches helps track the health decline accurately; nonetheless, we hypothesize that it needs to dynamically incorporate the evolution of psychosocial factors as the older adults age while parameterizing the neurodegenerative disease progression as an indicator of overall health decline. The functional health changes (e.g. wandering, incorrect sequence, etc,) manifested through behavioral symptoms (agitation, depression) during the early stages of dementia are subtle but still can be quantified through the interruption and deviation from specific activities. The stages of dementia with respect to cognitive impairments e.g., no, mild, moderate, and severe cognitive decline are labeled across the distributions of data conditioned on the user using survey and observation-based assessments in a clinical setting as opposed to the user’s current and past activities. The latter is practically expensive due to the challenges in longitudinal sensing and data collection including the appropriate labeling of the data. To address this challenge, we propose a reinforcement learning-based framework that leverages the placement of the representations learned via contrastive loss and the clinician input to model a reward function that enables us to integrate the psychosocial factors with the neurodegenerative disorder indicator. We evaluated our framework on 25 older adults dataset collected from a retirement community center with IRB approval in place. Our proposed self-taught learning approach reduces the activity missed detection by 20%, the contrastive learning approach obtains an F1-score of 92% in detecting demarcation of overlapped activities, and the multi-task learning approach achieves an F1-score of 95% in classifying cognitive impairments concurrently with activities. Next, we plan to evaluate our reinforcement learning approach using our own and public datasets. Finally, we plan to build a mobile application, that connects with smart home IoT devices, harnesses data, executes the proposed algorithms for health assessments, and provides just-in-time feedback about the current cognitive health status to both the user and clinicians along with the explanation for the recommended assessment.

 

Announcement of Ph.D. Proposal Defense

Title: Towards Understanding Usable Privacy Concerns Among Older Adults
Student: Hirak Ray
Date/Time: Friday May 14th, 2021, 10am-1pm Eastern Time
Linkhttps://umbc.webex.com/umbc/j.php?MTID=mfdc0954b21f066b479a6f30af9ac2e28
Committee:
Dr. Ravi Kuber, UMBC (Co-Chair)
Dr. Adam J. Aviv, George Washington University (Co-Chair), (aaviv@gwu.edu)
Dr. Foad Hamidi, UMBC
Dr. Yaxing Yao, UMBC
Dr. Florian Schaub, University of Michigan (fschaub@umich.edu)

Abstract

While researchers have examined privacy among a wide range of users, there has been lesser focus on the perceptions of older adults, who may exhibit attitudes towards privacy that differs from their younger counterparts. Lower levels of awareness regarding potential online privacy violations, coupled with limited knowledge of protective measures that can be adopted to counter online attacks lead to negative outcomes for older adults, including falling victim to scams and data breaches.

My research investigates older adults’ privacy and security perceptions regarding digital and non-digital technologies, and identifies how their usage and adoption of tools and technologies are impacted by these perceptions. To this end, two studies have been conducted, exploring (1) complex privacy behaviors of older adults and comparisons with younger age groups, and (2) adoption barriers and motivators of online security tools.

Currently, older adults are experiencing a higher amount of social isolation than before due to the social distancing requirements of the COVID-19 lockdown. For my dissertation, I intend to investigate the forced adoption of online tools among older adults due to this phenomenon, specifically towards online privacy and security issues which may surface, and see how this compares to adoption under less constrained circumstances. Findings from the work would complement findings from my prior studies to aid design guidance provided to interface developers and researchers aiming to support privacy among older groups of users.

 

Announcement of Ph.D. Dissertation Defense

Title: Investigating Mental Models Of Risk Among Security-Expert Users
Student: Flynn Wolf
Date/time: Monday, May 17th, 2021, 12pm-3pm Eastern Time

Committee:

Dr. Ravi Kuber, UMBC (Co-Chair)
Dr. Adam J. Aviv, George Washington University (Co-Chair), (aaviv@gwu.edu)
Dr. Foad Hamidi, UMBC
Dr. Anita Komlodi, UMBC
Dr. Michelle Mazurek, University of Maryland College Park (mmazurek@cs.umd.edu)

Abstract

Usable security research endeavors to make interaction with technology more intuitive and trustworthy. Studying this area is critical given the increasing amount of sensitive data we trust to networked devices and services, and our general propensity to discard online safeguards we find burdensome. A valuable facet of this research is qualitative study of users’ mental models. This approach gathers description and observation of user behavior, and creates an interpretive picture of how users understand their technology. Users’ experiences with their mobile devices shape these beliefs, and that constructed understanding in turn may profoundly influence users’ technological choices and expectations. Security-expert users are a particularly interesting cohort because experience may sensitize them to risk and guide their technology choices in unique or prescient ways. With these issues in mind, three qualitative inquiries have been made to better understand how advanced mental models of network security may influence user behavior.

First, we investigated experts’ (n=20) understanding of mobile security and how their concerns shaped everyday use of those platforms. They experienced typical usability problems and situational impairments with their mobile devices, but also described caution towards data sharing. The avoidance was based on factors including the sensitivity of the data to be shared and variable distrust of underlying network connections and technology platforms. We then secondly compared the outlook and behavior of a similar cohort (n=38) of both experts and non-experts when considering adoption of biometric authentication on their mobile devices. Experts were found to more readily accept fingerprint unlocking as an improved mode of authentication than non-experts. However, experts resisted its use for sensitive transactions, and their enthusiasm did not transfer to facial recognition. Our third study examined perceptions of experts, in two rounds (n=8, and n=19), serving security consultant roles (CISOs) for small businesses. CISOs tended to view government-authored security guidance as harder to use but more authoritative than commercial sources, and saw small businesses as highly vulnerable to online threats.

This work concludes by comparing the observations and implications drawn from these studies to examples of security guidance. Guidance of this type was described as a key reference and basis for advisement by CISOs to non-expert small business owners. Based on the deductive comparison, a preliminary set of guidelines for presenting models of online risk are offered. These guidelines indicate ways to make security guidance more effective, based on experts’ perspectives that should be broadly applicable and useful as awareness of online threats becomes more prevalent.

 

Recent Defenses

Title: Investigating Ways to Support the Use of Voice Assistants (VAs) Among Individuals Who Are Blind
Student: Ali Abdolrahmani
Date/time: Friday, April 16th, 2021, 1pm-4pm
Committee:
Dr. Ravi Kuber, UMBC (Co-Chair)
Dr. Stacy Branham, University of California Irvine (Co-Chair), (sbranham@uci.edu)
Dr. Anita Komlodi, UMBC
Dr. Foad Hamidi, UMBC
Dr. Shiri Azenkot, Cornell Tech, (shiri.azenkot@cornell.edu)
Abstract
Voice assistants (VAs) (e.g., Apple Siri, Google Assistant, Amazon Alexa, etc.) offer considerable potential to users. The hands-free nature of these technologies, enables users to perform tasks which would otherwise be difficult to accomplish. As individuals who are blind are already familiar with spoken output through the use of screen readers, it is surmised that mainstream voice interfaces (available on different VA platforms and applications) may serve as a promising modality to support tasks in a variety of settings.My dissertation research draws upon two bodies of work: (1) understanding the interaction experiences of individuals who are blind using VAs, and (2) examining ways these mainstream technologies can support indoor navigation and their information needs. I aim to investigate how VAs can be extended to enhance the indoor navigation experience through airports.  These venues present interesting navigational challenges to travelers (e.g., large unfamiliar space, presence of noise and obstacles etc.).  The findings of this research contribute to addressing the indoor navigation challenges for individuals who are blind. In addition, there is potential for findings to address the needs of the general population, who may also experience difficulties navigating within these settings.

 

Title: Exploring the Relationship between Body Expressions and Electrodermal Activity in Public Speaking Anxiety
Student: Shu Yi Tseng
Date/time: Monday, April 12th, 2021, 10pm-12pm
Link: Virtual
Committee:
Dr. Andrea Kleinsmith, UMBC (Chair)
Dr. Ravi Kuber, UMBC
Dr. Anita Komlodi, UMBC
Abstract
Over the last decade, a significant amount of research has been dedicated to understanding public speaking anxiety (PSA) and building systems for practicing public speaking skills. Input to these systems is often multimodal, focusing on verbal and nonverbal communication, including conversational gestures, vocal characteristics, gaze behavior, and physiological responses (such as electrodermal activity (EDA)). However, it is unclear how, when and why some behaviors manifest. That is, if we can determine potential behaviors that contribute to elevating or easing the speaker’s anxiety level, this information can be integrated into an interactive virtual audience within a training platform for public speaking and passed on to the speaker as individualized feedback. Reflecting on this feedback may help those suffering from PSA learn to better mitigate their PSA by gaining insight and awareness into their PSA-related expressions as a step toward learning how to employ potential coping strategies. Inspired by Lee & Kleinsmith’s previous work, the goal of this thesis is to explore the relationship between EDA and nonverbal behaviors and how information from these modalities can be implemented into the design of a public speaking training platform. We applied unsupervised K-means clustering to explore how students were grouped according to body and EDA features only. Then, ordinal logistic regression was implemented to measure the contribution of each feature in predicting audience-perceived PSA of the speakers. The results of each model are compared and contrasted, leading to a set of guidelines for designing a public speaking training platform.

 

Title: Understanding Immersion Experience when Using Virtual Reality for Foreign Language Learning
Student: Hye-Kyung Bae (Ph.D. student, Human-Centered Computing)
Date/time: Tuesday 19th, January 2021, 11am-1pm
Committee:
Dr. Wayne Lutters, Information Systems, UMBC (Co-Chair)
Dr. Anita Komlodi, Information Systems, UMBC (Co-Chair)
Dr. Ravi Kuber, Information Systems, UMBC
Dr. Zane Berge, Department of Education, UMBC
Dr. Kyung-Eun Yoon, Modern Languages, Linguistics & Intercultural Communication, UMBC
Abstract
One of the most effective ways to learn a foreign language is an immersion in its native cultures. While visiting another country, immersion experiences accelerate learning and confidence by dining in restaurants, shopping in markets, participating in events, and meeting new people. However, in practice, few students have access to these kinds of experiences. Virtual Reality (VR) technology has the potential to expand access allowing more students to have similar immersion experiences. These experiences are more engaging than conventional methods such as passively watching videos or interacting with two-dimensional user interfaces. However, are they more conducive to learning?
Some recent studies have dealt with VR for foreign language learning, which includes either artificial graphics-based VR simulations or naturalistic 360-degree VR videos. However, few studies have compared these two different approaches. In VR simulations participants learn vocabulary by interacting with artificial objects or agents. In contrast, 360-degree videos provide a more real-life experience, but with fewer opportunities for interaction. Both support and encourage independent movement inspiring learners to freely explore the virtual world on their own to improve their language skills in context.
This dissertation project includes three studies to understand immersion experiences in different VR conditions for foreign language learning. Study0 as an initial study to provide proof of concept for this approach was carried out using 360-degree videos with guided components. Study1 was conducted using two sets of 360-degree videos of real-life settings, comparing unguided vs. guided components. Study2 will compare 360-degree videos to VR simulations. Initial results from Study0 and Study1 have been promising, such as performing better for learning with the guidance and exploring more on their own without the guidance. The expected contributions are to provide a better understanding of students’ VR experience for foreign language learning, some guidance to people who would choose VR applications, and design implications and suggestions for those who would create VR content.

 

Title: Applying Virtual Reality to Motivate and Prepare International Students for Community Engagement
Student: Iman Alshathri (Ph.D. IS)
Date/time: Tuesday 12th January 2021, 10am-12pm
Committee:
Dr. Anita Komlodi (Co-Chair)
Dr. Ravi Kuber (Co-Chair)
Dr. Foad Hamidi
Dr. Lina Zhou, UNCC (lzhou8@uncc.edu)
Dr. Norah Abokhodair, Microsoft Learning Innovation Studio (norah.abokhodair@microsoft.com)
Abstract
There are numerous benefits to students participating in community engagement activities.   However, levels of participation among international students are lower compared to their domestic counterparts. Virtual reality has been shown to have the potential to encourage behavioral change among individuals in a range of contexts.  Researchers have yet to examine the impact of virtual reality to promote community engagement among international students. Our preliminary investigations suggest that this technology offers promise to motivate and prepare international graduate students to volunteer, by supporting the changing of perceptions and attitudes towards volunteering to support community engagement initiatives. More specifically, the aim of this research is to examine the role virtual reality technology plays in encouraging international students to volunteer in a community that is different from their own.  Two studies are proposed as part of this work.  Findings will lead to the development of suggestions to support community engagement trainers who are interested in applying virtual reality to motivate and prepare individuals to volunteer.

 

Title: Understanding and Supporting Collaboration Between Youth in Technical Workplaces
Student: William Easley
Time: Friday, September 11, 2020, 12 pm – 3 pm
Location: Meeting Link: https://umbc.webex.com/umbc/j.php?MTID=mc3c17c758cbb6d0a640de365bf203ccc
Meeting number (access code): 120 111 8670
Meeting password: 3exRkqNYu34

Committee: 
Dr. Foad Hamidi (Chair/Advisor)
Dr. Helena Mentis (Chair/Advisor)
Dr. Anita Komlodi
Dr. Amy Hurst
Dr. Wayne G. Lutters
Dr. Betsy DiSalvo

Abstract:
Emerging digital technological advances have been a driving force in the evolution of the modern workplace. This influence extends not only to the creation of new industries and types of work, but also to how work is completed. For example, new technologies have made it possible for us to coordinate and communicate over vast geographic distances. In response to these changes, considerable efforts have been made to re-train our existing workforce to have the technical and communi- cation skills necessary to be successful. This shifting landscape presents exciting opportunities to focus on ways to support youth – who have spent their entire lives with access to inter-networked technologies – as they prepare to enter the workplace of the future.
This dissertation describes longitudinal research conducted at the Digital Harbor Foundation 3D Print Shop over a period of more than three years. The print shop offers after-school technical employment to youth from diverse backgrounds. To understand the experiences and challenges faced by this population as they transitioned into this environment, we have utilized a variety of methods including participant and direct observation, interviews, the analysis of chat logs, a focus group, and a participatory elicitation method.

The three studies included in this document all seek to better understand how youth coordinate work in technical settings with a specific focus on their usage of communication technologies. The first documents and unpacks instances where youth successfully and unsuccessfully coordinated their work during asynchronous handoffs. The second study characterizes the youths’ use of Slack, a popular workplace chatting tool and investigates the factors motivating and limiting adoption of this tool. The final study examines the youths’ attitudes and perceptions towards workplace communication technologies. Findings and recommendations from this dissertation offer insight into how we can better prepare and support our future workforce.

Title: Interpretable deep learning models for Electronic Health Records
Student: Peichang Shi
Advisor: Dr. Aryya Gangopadhyay
Time: Monday, June 22, 2019, 10 am – 12 pm
Location: Virtual

Abstract:
Deep learning neural networks (DNNs) have been proved to be powerful tools in computer vision and machine learning across robotics, healthcare, and security. However, despite their superior performance to traditional machine learning methods, it remains challenging to understand their inner mechanism and explain their downstream output due to the black box effects.

In this dissertation proposal, we aim to unfold the black-box and provide explanations in a transparent way using Electronic Health Records (EHRs). Our proposed approach is to add a partially connected layer after the output from the DNNs based on a decision tree. This decision tree layer could interpret DNNs through rules from the input vectors. Our method could achieve global interpretability for all the test instances on the hidden layers without sacrificing the accuracy obtained from the original deep learning model. This means our model is faithful to the original deep DNN model, which leads to reliable interpretations. Beyond that, we would also like to address some common issues related to EHRs including irregular time series data, censored data, multitask learning issues and hierarchical structure variables in the model.

The methods proposed in this dissertation will be validated with the End Stage Renal Disease (ESRD) Population Public Use File (PUF). This PUF data contains the current and historical ESRD patient population and associated clinical data, which has about 3.2 million patients including hospital admissions, hospitalizations, chronic conditions and 13 monthly clinical measures.