UMD Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/3
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
4 results
Search Results
Item DOCUMENT INFORMATION EXTRACTION, STRUCTURE UNDERSTANDING AND MANIPULATION(2023) Mathur, Puneet; Manocha, Dinesh; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Documents play an increasingly central role in human communications and workplace productivity. Every day, billions of documents are created, consumed, collaborated on, and edited. However, most such interactions are manual or rule-based semi-automated. Learning from semi-structured and unstructured documents is a crucial step in designing intelligent systems that can understand, interpret, and extract information contained in digital PDFs, forms, receipts, contracts, infographics, etc. Our work tries to solve three major problems in the domain of information extraction from real-world multimodal (text+images+layout) documents: (1) multi-hop reasoning between concepts and entities spanning several paragraphs; (2) semi-structured layout extraction in documents consisting of thousands of text tokens and embedded images arranged in specific layouts; (3) hierarchical document representations and the need to transcend content lengths beyond a fixed window for effective semantic reasoning. Our research broadly binds together the semantic (document-level information extraction) and structural (document image analysis) aspects of document intelligence to advance user productivity. The first part of the research addresses issues related to information extraction from characteristically long-range documents that consist of multiple paragraphs and require long-range contextualization. We propose augmenting the capabilities of the Transformer-based methods with graph neural networks to capture local-level context as well as long-range global information to solve document-level information extraction tasks. In this aspect, we first solve the task of document-level temporal relation extraction by leveraging rhetorical discourse features, temporal arguments, and syntactic features through a Gated Relational-GCN model to extend the capability of Transformer architecture for discourse-level modeling. Next, we propose DocTime, a novel temporal dependency graph parsing method that utilizes structural, syntactic, and semantic relations to learn dependency structures over time expressions and event entities in text documents to capture long-range interdependencies. We also show how the temporal dependency graphs can be incorporated into the self-attention layer of Transformer models to improve the downstream tasks of temporal questions answering and temporal NLI. Finally, we present DocInfer - a novel, end-to-end Document-level Natural Language Inference model that builds a hierarchical document graph, performs paragraph pruning, and optimally selects evidence sentences to identify the most important context sentences for a given hypothesis. Our evidence selection mechanism allows it to transcend the input length limitation of modern BERT-like Transformer models while presenting the entire evidence together for inferential reasoning that helps it to reason on large documents where the evidence may be fragmented and located arbitrarily far apart. The second part of the research covers novel approaches for understanding, manipulation, and downstream applications of spatial structures extracted from digital documents. We first propose LayerDoc to extract the hierarchical layout structure in visually rich documents by leveraging visual features, textual semantics, and spatial coordinates along with constraint inference in a bottom-up layer-wise fashion. Next, we propose DocEditor, a Transformer-based localization-aware multimodal (textual, spatial, and visual) model that performs the novel task of language-guided document editing based on user text prompts. Further, we investigated methods for building text-to-speech systems for semi-structured documents. Finally, we will explore two applications of long-context document-level reasoning: (i) user-personalized speech recognition systems for improved next-word prediction in specific domains by utilizing retrieval augmentation techniques for ASR Language Models; (ii) Transformer-based methods to utilize multimodal information from long-form financial conference calls (document-level transcripts, audio-visual recordings, and tabular information) for improved financial time series prediction tasks.Item ASSESSING QUALITY IN HIGH-UNCERTAINTY MARKETS: ONLINE REVIEWS OF CREDENCE SERVICES(2016) Lantzy, Shannon; Stewart, Katherine; Viswanathan, Siva; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)In economics of information theory, credence products are those whose quality is difficult or impossible for consumers to assess, even after they have consumed the product (Darby & Karni, 1973). This dissertation is focused on the content, consumer perception, and power of online reviews for credence services. Economics of information theory has long assumed, without empirical confirmation, that consumers will discount the credibility of claims about credence quality attributes. The same theories predict that because credence services are by definition obscure to the consumer, reviews of credence services are incapable of signaling quality. Our research aims to question these assumptions. In the first essay we examine how the content and structure of online reviews of credence services systematically differ from the content and structure of reviews of experience services and how consumers judge these differences. We have found that online reviews of credence services have either less important or less credible content than reviews of experience services and that consumers do discount the credibility of credence claims. However, while consumers rationally discount the credibility of simple credence claims in a review, more complex argument structure and the inclusion of evidence attenuate this effect. In the second essay we ask, “Can online reviews predict the worst doctors?” We examine the power of online reviews to detect low quality, as measured by state medical board sanctions. We find that online reviews are somewhat predictive of a doctor’s suitability to practice medicine; however, not all the data are useful. Numerical or star ratings provide the strongest quality signal; user-submitted text provides some signal but is subsumed almost completely by ratings. Of the ratings variables in our dataset, we find that punctuality, rather than knowledge, is the strongest predictor of medical board sanctions. These results challenge the definition of credence products, which is a long-standing construct in economics of information theory. Our results also have implications for online review users, review platforms, and for the use of predictive modeling in the context of information systems research.Item PRINCIPLES OF INFORMATION PROCESSING IN NEURONAL AVALANCHES(2011) Yang, Hongdian; Roy, Rajarshi; Plenz, Dietmar; Chemical Physics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)How the brain processes information is poorly understood. It has been suggested that the imbalance of excitation and inhibition (E/I) can significantly affect information processing in the brain. Neuronal avalanches, a type of spontaneous activity recently discovered, have been ubiquitously observed in vitro and in vivo when the cortical network is in the E/I balanced state. In this dissertation, I experimentally demonstrate that several properties regarding information processing in the cortex, i.e. the entropy of spontaneous activity, the information transmission between stimulus and response, the diversity of synchronized states and the discrimination of external stimuli, are optimized when the cortical network is in the E/I balanced state, exhibiting neuronal avalanche dynamics. These experimental studies not only support the hypothesis that the cortex operates in the critical state, but also suggest that criticality is a potential principle of information processing in the cortex. Further, we study the interaction structure in population neuronal dynamics, and discovered a special structure of higher order interactions that are inherent in the neuronal dynamics.Item Attachment Security and the Processing of Attachment-Relevant Social Information in Late Adolescence(2006-04-26) Dykas, Matthew Jason; Cassidy, Jude A; Psychology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)According to attachment theory, internal working models of attachment function to influence the ways in which individuals obtain, organize, and operate on attachment-relevant social information (Bowlby, 1980). The principal aim of this investigation was the examination of whether adolescents' internal working models of attachment are linked to their memory for attachment-relevant social information. I proposed that adolescents who possess negative internal working models of attachment (i.e., insecure adolescents and adolescents who possess negative representations of their parents) process attachment-relevant social information differently from adolescents who possess positive internal working models of attachment (i.e., secure adolescents and adolescents who possess positive representations of their parents). I also proposed that such differences are associated with two distinct patterns of attachment-relevant social information-processing. More precisely, I hypothesized that insecure adolescents and adolescents who possess negative representations of their parents are more likely to <em>suppress</em> attachment-relevant social information (from entering conscious awareness) in some circumstances, and to process attachment-relevant social information in a <em>negatively-biased schematic manner</em> in others. To test this hypothesis, I tapped adolescents' (n = 189) internal working models of attachment by assessing their "state of mind with respect to attachment" (as assessed using the Adult Attachment Interview), representations of parents, and attachment-related romantic anxiety and avoidance (as assessed using the Experiences in Close Relationships Inventory). I used four experimental tasks to assess adolescents' memory for attachment-relevant social information. Many of the findings reported in this investigation can be viewed as supporting the notion that insecure adolescents and adolescents who possess negative representations of their parents either suppress attachment-relevant social information or process such information in a negatively-biased schematic manner. For example, in the experimental task that tapped suppression, insecure adolescents showed poorer memory for emotionally-significant childhood experiences. Moreover, in all three of the experimental tasks tapping schematically-driven social information-processing, insecure adolescents and adolescents who possessed negative representations of their parents showed either greater memory for negative parental attributes or more negative reconstructive memory for conflict. In addition to these principal findings, evidence emerged that adolescent attachment was linked to memory for peer-related information, as well as to parents' reconstructive memory for adolescent-parent conflict.