Computer Science Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/2756
Browse
15 results
Search Results
Item Developing and Measuring Latent Constructs in Text(2024) Hoyle, Alexander Miserlis; Resnik, Philip; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Constructs---like inflation, populism, or paranoia---are of fundamental concern to social science. Constructs are the vocabulary over which theory operates, and so a central activity is the development and measurement of latent constructs from observable data. Although the social sciences comprise fields with different epistemological norms, they share a concern for valid operationalizations that transparently map between data and measure. Economists at the US Bureau of Labor Statistics, for example, follow a hundred-page handbook to sample the egg prices that constitute the Consumer Price Index; Clinical psychologists rely on suites of psychometric tests to diagnose schizophrenia. In many fields, this observable data takes the form of language: as a social phenomenon, language data can encode many of the latent social constructs that people care about. Commensurate with both increasing sophistication in language technologies and amounts of available data, there has thus emerged a "text-as-data" paradigm aimed at "amplifying and augmenting" the analyses that compose research. At the same time, Natural Language Processing (NLP), the field from which analysis tools originate, has often remained separate from real-world problems and guiding theories---as least when it comes to social science. Instead, it focuses on atomized tasks under the assumption that progress on low-level language aspects will generalize to higher-level problems that involve overlapping elements. This dissertation focuses on NLP methods and evaluations that facilitate the development and measurement of latent constructs from natural language, while remaining sensitive to social sciences' need for interpretability and validity.Item ANALYZING COMMUNICATIVE CHOICES TO UNDERSTAND THEIR MOTIVATIONS, CONTEXT-BASED VARIATION, AND SOCIAL CONSEQUENCES(2023) Goel, Pranav; Resnik, Philip; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)In many settings, communicating in a language requires making choices among different possibilities — the issues to focus on, the aspects to highlight within any issue, the narratives to include, and more. These choices, deliberate or not, aresocially structured. The ever-increasing availability of unstructured large-scale textual data, in part due to the bulk of communication and information dissemination happening in online or digital spaces, makes natural language processing (NLP) techniques a natural fit for helping understand socially-situated communicative choices using that textual data. Within NLP methods, unsupervised NLP methods are often needed since digital large-scale textual data in the wild is often available without accompanying labels, and any existing labels or categorization might not be appropriate for answering specific research questions. This dissertation seeks to address the following question: how can we use unsupervised NLP methods to study texts authored by specific people or institutions in order to effectively explicate the communicative choices being made, as well as to investigate their potential motivations, context-based variation, and consequences? Our first set of contributions centers on methodological innovation. We focus on topic modeling: a class of generally unsupervised NLP methods that can automatically discover authors’ communicative choices in the form of topics or categorical themes present in a collection of documents. We introduce a new neural topic model (NTM) that effectively incorporates contextualizing sequential knowledge. Next, we find critical gaps in the near-universal automated evaluation paradigm that compares different models in the topic modeling methods literature, which calls into question much of the recent work in NTM development claiming “state-of-the-art” and emphasizes the importance of validating the outputs of unsupervised NLP methods. In order to use unsupervised NLP methods to investigate potential motivations, context-based variation, and consequences of communicative choices, we link textual data with information about the authors, social contexts, and media involved in their production — these connected information sources help us conduct empirical research in social sciences. In our second set of contributions, we analyze a previously unexplored connection between a politician’s donors and their communicative choices in their floor speeches to show how donations influence issue-attention in US Congress, enabling a new look at money in politics and providing an example of studying motivations behind communicative choices. Our third set of contributions uses text-based ideal points to better understand the role of institutional constraints and audience considerations in the varying expression and ideological positioning of politicians. The application of this tool for expanding knowledge of legislative politics is enabled by comprehensive annotations for modeling outputs provided by domain experts in order to establish the tool’s validity and reliability. In our fourth set of contributions, we demonstrate the potential of both unsupervised NLP techniques and social network data and methods in better understanding the downstream consequences of communicative choices. We focus on misinformation narratives in mainstream media, viewing and highlighting misinformation as something that goes beyond just false claims published by certain bad actors or stories published by certain ‘fake news’ outlets. Our findings suggest a strategic repurposing of mainstream news by conveyors of misinformation as a way to enhance the reach and persuasiveness of misleading narratives.Item Design Considerations for Remote Expert Guidance Using Extended Reality in Skilled Hobby Settings.(2023) Maddali, Hanuma Teja; Lazar, Amanda; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)As compact and lightweight extended reality (XR) devices become increasingly available, research is being reinvigorated in a number of areas. One such area for XR applications involves remote collaboration, where a remote expert can assist, train, or share skills or ideas with a local user to solve a real-world task. For example, researchers have looked into real-time expert assistance and professional training of novices in skilled physical activities such as field servicing and surgical training. Even as our understanding of XR for remote collaboration in professional settings advances, an area that has not been examined is how XR can support such expert-novice collaboration in skilled hobby activities (e.g., gardening, woodworking, and knitting). Metrics such as task accuracy or efficiency are often less important than in professional settings. Instead, other dimensions, such as social connectedness and emotional experience, may become central dimensions that inform system design. In my dissertation, I examine how the XR environment can be designed to support the sharing of skills in hobby activities. I have selected gardening as a hobby activity to examine remote skill-sharing in XR between experts and novices. Like in other hobby activities, learning gardening practices remotely can involve asynchronous, text, or image/video-based communication on Facebook groups. While these may be helpful for individual questions, they do not capture the social, affective, and embodied dimensions of gaining expertise as a novice through situated learning in the garden. These dimensions can also be central to the experience of the activity. In my work, I seek to understand how to design a social XR environment that captures these dimensions in ways that are acceptable and useful to intergenerational expert-novice gardener groups. Through my dissertation work, I answer the following research questions:1. How do practitioners of a particular hobby exhibit sociality and what kinds of social interactions facilitate skill-sharing? What are some key opportunities for computer-supported collaborative work in this space? 2. What are practitioners' perceptions of using XR for skill-sharing? What are the important dimensions of the design space and design scenarios for social XR systems? 3. How do practitioners use different components of the activity space (e.g., tools or sensory stimuli) and their affordances to facilitate social connection? What context is essential to capture when reconstructing these objects virtually for remote interaction in XR (e.g., interactivity and realism)? 4. What are some design considerations for XR to support accessible interactions that reflect the values and goals of an intergenerational group?Item Data-driven Storytelling in Dynamic Graph Comics through Hierarchical Clustering(2023) Kannan, Abhinav; Elmqvist, Niklas; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)In this work, we propose a tool to generate a dynamic graph comic given dense, time-series edge-vertex data. Prior research has demonstrated the effectiveness of node-link diagrams as an expressive medium for storytelling with dynamic graphs, and in this work, we develop an interface that generates a customizable comic strip consisting of node-link diagram snapshots. We use hierarchical aggregation to cluster and pile graphs based on the number of frames a user may wish to see, with each frame depicting a snapshot in time. We validate the interface with real-world datasets to understand temporal changes in a graph network, and evaluate the interface against an expert audience. Finally, we propose a path forward for improvement of dynamic graph comics as a storytelling medium.Item Algorithms and Data Structures for Faster Nearest-Neighbor Classification(2022) Flores Velazco, Alejandro; Mount, David; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Given a set P of n labeled points in a metric space (X,d), the nearest-neighbor rule classifies an unlabeled query point q ∈ X with the class of q's closest point in P. Despite the advent of more sophisticated techniques, nearest-neighbor classification is still fundamental for many machine-learning applications. Over the years, this~has motivated numerous research aiming to reduce its high dependency on the size and dimensionality of the data. This dissertation presents various approaches to reduce the dependency of the nearest-neighbor rule from n to some smaller parameter k, that describes the intrinsic complexity of the class boundaries of P. This is of particular significance as it is usually assumed that k ≪ n on real-world training sets. One natural way to achieve this dependency reduction is to reduce the training set itself, selecting a subset R ⊆ P to be used by the nearest-neighbor rule~to~answer incoming queries, instead of using P. Evidently, this approach would reduce the dependencies of the nearest-neighbor rule from n, the size of P, to the size of R. This dissertation explores different techniques to select subsets whose sizes are proportional to k, and that provide varying degrees of correct classification guarantees. Another alternative involves bypassing training set reduction, and instead building data structures designed to answer classification queries directly. To this end, this dissertation proposes the Chromatic AVD; a Quadtree-based data structure designed to answer ε-approximate nearest-neighbor classification queries. The query time and space complexities of this data structure depend on k_ε; a generalization of k that describes the intrinsic complexity of the ε-approximate class boundaries of P.Item TOWARDS AUTONOMOUS DRIVING IN DENSE, HETEROGENEOUS, AND UNSTRUCTURED TRAFFIC(2022) Chandra, Rohan; Manocha, Dinesh; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation addressed many key problems in autonomous driving towards handling dense, heterogeneous, and unstructured traffic environments. Autonomous vehicles (AV) at present are restricted to operating on smooth and well-marked roads, in sparse traffic, and among well-behaved drivers. We developed new techniques to perceive, predict, and plan among human drivers in traffic that is significantly denser in terms of number of traffic-agents, more heterogeneous in terms of size and dynamic constraints of traffic agents, and where many drivers do not follow the traffic rules. In this thesis, we present work along three themes—perception, driver behavior modeling, and planning. Our novel contributions include: 1. Improved tracking and trajectory prediction algorithms for dense and heterogeneous traffic using a combination of computer vision and deep learning techniques. 2. A novel behavior modeling approach using graph theory for characterizing human drivers as aggressive or conservative from their trajectories. 3. Behavior-driven planning and navigation algorithms in mixed (human driver and AV) and unstructured traffic environments using game theory and risk-aware control. Additionally, we have released a new traffic dataset, METEOR, which captures rare and interesting, multi-agent driving behaviors in India. These behaviors are grouped into traffic violations, atypical interactions, and diverse scenarios. We evaluate our perception work on tracking and trajectory prediction using standard autonomous driving datasets such as the Waymo Open Motion, Argoverse, NuScenes datasets, as well as public leaderboards where our tracking approach resulted in achieving rank 1 among over a 100 methods. We apply human driver behavior modeling in planning and navigation at unsignaled intersections and highways scenarios using state-of-the-art traffic simulators and show that our approach yields fewer collisions and deadlocks compared to methods based on deep reinforcement learning. We conclude the presentation with a discussion on future work.Item Exploring Blind and Sighted Users’ Interactions With Error-Prone Speech and Image Recognition(2021) Hong, Jonggi; Kacorri, Hernisa; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Speech and image recognition, already employed in many mainstream and assistive applications, hold great promise for increasing independence and improving the quality of life for people with visual impairments. However, their error-prone nature combined with challenges in visually inspecting errors can hold back their use for more independent living. This thesis explores blind users’ challenges and strategies in handling speech and image recognition errors through non-visual interactions looking at both perspectives: that of an end-user interacting with already trained and deployed models such as automatic speech recognizer and image recognizers but also that of an end-user who is empowered to attune the model to their idiosyncratic characteristics such as teachable image recognizers. To better contextualize the findings and account for human factors beyond visual impairments, user studies also involve sighted participants on a parallel thread. More specifically, Part I of this thesis explores blind and sighted participants' experience with speech recognition errors through audio-only interactions. Here, the recognition result from a pre-trained model is not being displayed; instead, it is played back through text-to-speech. Through carefully engineered speech dictation tasks in both crowdsourcing and controlled-lab settings, this part investigates the percentage and type of errors that users miss, their strategies in identifying errors, as well as potential manipulations of the synthesized speech that may help users better identify the errors. Part II investigates blind and sighted participants' experience with image recognition errors. Here, we consider both pre-trained image recognition models and those fine-tuned by the users. Through carefully engineered questions and tasks in both crowdsourcing and semi-controlled remote lab settings, this part investigates the percentage and type of errors that users miss, their strategies in identifying errors, as well as potential interfaces for accessing training examples that may help users better avoid prediction errors when fine-tuning models for personalization.Item Search Among Sensitive Content(2021) Sayed, Mahmoud; Oard, Douglas W.; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Current search engines are designed to find what we want. But many collections can not be made available for search engines because they contain sensitive content that needs to be protected. Before release, such content needs to be examined through a sensitivity review process, which can be difficult and time-consuming. To address this challenge, search technology should be capable of providing access to relevant content while protecting sensitive content. In this dissertation, we present an approach that leverages evaluation-driven information retrieval (IR) techniques. These techniques optimize an objective function that balances the value of finding relevant content with the imperative to protect sensitive content. This requires evaluation measures that balance between relevance and sensitivity. Baselines are introduced for addressing the problem, and a proposed approach that is based on building a listwise learning to rank model is described. The model is trained with a modified loss function to optimize for the evaluation measure. Initial experiments re-purpose a LETOR benchmark dataset, OHSUMED, by using Medical Subject Heading (MeSH) labels to represent the sensitivity. A second test collection is based on the Avocado Research Email Collection. Search topics were developed as a basis for assessing relevance, and two personas describing the sensitivities of representative (but fictional) content creators were created as a basis for assessing sensitivity. These personas were based on interviews with potential donors of historically significant email collections and with archivists who currently manage access to such collections. Two annotators then created relevance and sensitivity judgments for 65 topics for one or both personas. Experiment results show the efficacy of the learning to rank approach. The dissertation also includes four extensions to increase the quality of retrieved results with respect to relevance and sensitivity. First, the use of alternative optimization measures is explored. Second, transformer-based rankers are compared with rankers based on hand-crafted features. Third, a cluster-based replacement strategy that can further improve the score of our evaluation measures is introduced. Fourth, a policy that truncates the ranked list according to the query's expected difficulty is investigated. Results show improvements in each case.Item Enabling On-body Computing Using a Track-Based Wearable(2021) Sathya Sai Kumar, Anup; Peng, Huaishu; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)We are seeing an increasing trend in the number of computing devices attached to the body which provide a myriad of data along with additional interaction mechanisms and haptic feedback from these locations. Although this provides more computing locations, the devices are still resigned to stay in those particular locations. We believe that relocatable wearables can reduce the number of devices that the user has to keep track of, while also providing dynamic data by moving around the body. Some attempts have been made to build relocatable wearables, but these attempts are either too bulky or make the use of unreliable and slow locomotion mechanisms. In this thesis, we present Calico, a miniature wearable robot system with fast and precise locomotion for on-body sensing, actuation, and interaction. Calico consists of a two-wheel robot and an on-cloth track system or "railway," on which the robot travels. The robot packs an IMU, a battery and an onboard microcontroller that supports wireless BLE communication. The track system has multiple branches that extend to key areas of the human body, allowing the robot to reach the front and back of the torso, shoulders and arms, and even lower body areas such as legs. The track system also includes rotational switches, or "railway turntables," enabling complex routing options when diverging tracks are presented. We introduce the system design of Calico and report a series of technical evaluations for system performance. To illustrate potential use cases, we present a suite of applications, including remote medical usage for chest auscultation, body posture tracking for training and exercise, a wearable walkie-talkie, and a robot companion living on-body.Item Operational challenges in dockless bike-shares: The case of hyperlocal Imbalance(2021) mishra, shivam; Paley, Derek DP; Systems Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Recent times have seen a shift from traditional docked to dockless bike-sharing systems. It is popular among consumers as it allows flexibility to drop off bikes anywhere and solves the last-mile problem of transportation. While convenient for users, the dockless bike-share system’s free-floating model introduces the problem of hyperlocal imbalance, about which little or no research is available. The hyperlocal imbalance is the supply-demand disparity created in a small geographical region due to consumer’s bias to pick up bikes from some locations compared to others. This paper introduces, demonstrates, and determines the reasons behind the hyperlocal-imbalance in dockless-bike-sharing systems. The study of hyperlocal imbalance requires access to fine-grained trip-level data, which is not easily accessible to the research community due to privacy or competition issues. To deal with it, in this work, we introduce an algorithm to extracttrip-level information from the General Bikeshare Feed Specification (GBFS) feeds, which bike-share companies are obliged to upload as per transportation department regulations across the US. The algorithm is validated against the actual trip data of dockless bikes. It extracts the trip details from the GBFS data with a recall of 77% and precision of 80%.