Theses and Dissertations from UMD

Permanent URI for this communityhttp://hdl.handle.net/1903/2

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 give thesis/dissertation in DRUM

More information is available at Theses and Dissertations at University of Maryland Libraries.

Browse

Subject: search.filters.subject.ontology

Search Results

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Semantic-driven modeling and reasoning for enhanced safety of cyber-physical systems
    (2016) Petnga, Leonard; Austin, Mark; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    This dissertation is concerned with the development of new methodologies and semantics for model-based systems engineering (MBSE) procedures for the behavior modeling of cyber-physical systems (CPS). Our main interest is to enhance system-level safety through effective reasoning capabilities embedded in procedures for CPS design. This class of systems is defined by a tight integration of software and physical processes, the need to satisfy stringent constraints on performance, safety and a reliance on automation for the management of system functionality. Our approach employs semantic–driven modeling and reasoning : (1) for the design of cyber that can understand the physical world and reason with physical quantities, time and space, (2) to improve synthesis of component-based CPS architectures, and (3) to prevent under-specification of system requirements (the main cause of safety failures in software). We investigate and understand metadomains, especially temporal and spatial theories, and the role ontologies play in deriving formal, precise models of CPS. Description logic-based semantics and metadomain ontologies for reasoning in CPS and an integrated approach to unify the semantic foundations for decision making in CPS are covered. The research agenda is driven by Civil Systems design and operation applications, especially the dilemma zone problem. Semantic models of time and space supported respectively by Allen’s Temporal Interval Calculus (ATIC) and Region Connectedness Calculus (RCC-8) are developed and demonstrated thanks to the capabilities of Semantic Web technologies. A modular, flexible, and reusable reasoning-enabled semantic-based platform for safety-critical CPS modeling and analysis is developed and demonstrated. The platform employs formal representations of domains (cyber, physical) and metadomains (temporal and spatial) entities using decidable web ontology language (OWL) formalisms. Decidable fragments of temporal and spatial calculus are found to play a central role in the development of spatio-temporal algorithms to assure system safety. They rely on formalized safety metrics developed in the context of cyber-physical transportation systems and collision avoidance for autonomous systems. The platform components are integrated together with Whistle, a small scripting language (under development) able to process complex datatypes including physical quantities and units. The language also enables the simulation, visualization and analysis of safety tubes for collision prediction and prevention at signalized and non-signalized traffic intersections.
  • Thumbnail Image
    Item
    SIMULTANEITY: On Urges and Origins
    (2014) Evans, Lauren Frances; Collis, Shannon; Art; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    All creative acts can be seen as intermediary attempts to bridge the gap between the known and the unknowable. In this thesis, I discuss a number of material and immaterial interfaces (e.g., the body, holes, desire, mankind, sacred sites) - all of which concern humanity and the divided nature of our existence in space, time, and matter. Prompted by urges and the allure of origins, my creative work addresses the body as a site of irresistible paradox. Inverting the boundaries between what is and what is not, it continually draws attention to the void of longing.
  • Thumbnail Image
    Item
    Semantic integration of geospatial concepts - a study on land use land cover classification systems
    (2011) Wei, Hua; Townshend, John; Geography; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    In GI Science, one of the most important interoperability is needed in land use and land cover (LULC) data, because it is key to the evaluation of LULC's many environmental impacts throughout the globe (Foley et al. 2005). Accordingly, this research aims to address the interoperability of LULC information derived by different authorities using different classificatory approaches. LULC data are described by LULC classification systems. The interoperability of LULC data hinges on the semantic integration of LULC classification systems. Existing works on semantically integrating LULC classification systems has a major drawback in finding comparable semantic representations from textual descriptions. To tackle this problem, we borrowed the method of comparing documents in information retrieval, and applied it to comparing LULC category names and descriptions. The results showed significant improvement comparing to previous works. However, lexical semantic methods are not able to solve the semantic heterogeneities in LULC classification systems: the confounding conflict - LULC categories under similar labels and descriptions have different LULC status in reality, and the naming conflict - LULC categories under different labels represent similar LULC type. Without confirmation of their actual land cover status from remote sensing, lexical semantic method cannot achieve reliable matching. To discover confounding conflicts and reconcile naming conflicts, we developed an innovative method of applying remote sensing to the integration of LULC classification systems. Remote sensing is a means of observation on actual LULC status of individual parcels. We calculated parcel level statistics from spectral and textural data, and used these statistics to calculate category similarity. The matching results showed this approach fulfilled its goal - to overcome semantic heterogeneities and achieve more reliable and accurate matching between LULC classifications in the majority of cases. To overcome the limitations of either method, we combined the two by aggregating their output similarities, and achieve better integration. LULC categories that post noticeable differences between lexical semantics and remote sensing once again remind us of semantic heterogeneities in LULC classification systems that must to be overcome before LULC data from different sources become interoperable and serve as the key to understanding our highly interrelated Earth system.
  • Thumbnail Image
    Item
    Patterns and Complexity in Biological Systems: A Study of Sequence Structure and Ontology-based Networks
    (2010) Glass, Kimberly; Girvan, Michelle; Physics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Biological information can be explored at many different levels, with the most basic information encoded in patterns within the DNA sequence. Through molecular level processes, these patterns are capable of controlling the states of genes, resulting in a complex network of interactions between genes. Key features of biological systems can be determined by evaluating properties of this gene regulatory network. More specifically, a network-based approach helps us to understand how the collective behavior of genes corresponds to patterns in genetic function. We combine Chromatin-Immunoprecipitation microarray (ChIP-chip) data with genomic sequence data to determine how DNA sequence works to recruit various proteins. We quantify this information using a value termed "nmer-association.'' "Nmer-association'' measures how strongly individual DNA sequences are associated with a protein in a given ChIP-chip experiment. We also develop the "split-motif'' algorithm to study the underlying structural properties of DNA sequence independent of wet-lab data. The "split-motif'' algorithm finds pairs of DNA motifs which preferentially localize relative to one another. These pairs are primarily composed of known transcription factor binding sites and their co-occurrence is indicative of higher-order structure. This kind of structure has largely been missed in standard motif-finding algorithms despite emerging evidence of the importance of complex regulation. In both simple and complex regulation, two genes that are connected in a regulatory fashion are likely to have shared functions. The Gene Ontology (GO) provides biologists with a controlled terminology with which to describe how genes are associated with function and how those functional terms are related to each other. We introduce a method for processing functional information in GO to produce a gene network. We find that the edges in this network are correlated with known regulatory interactions and that the strength of the functional relationship between two genes can be used as an indicator of how informationally important that link is in the regulatory network. We also investigate the network structure of gene-term annotations found in GO and use these associations to establish an alternate natural way to group the functional terms. These groups of terms are drastically different from the hierarchical structure established by the Gene Ontology and provide an alternative framework with which to describe and predict the functions of experimentally identified groups of genes.
  • Thumbnail Image
    Item
    Applications of Factorization Theorem and Ontologies for Activity ModelingRecognition and Anomaly Detection
    (2005-05-06) Akdemir, Umut; Chellappa, Rama; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    In this thesis two approaches for activity modeling and suspicious activity detection are examined. First is application of factorization theorem extension for deformable models in two dierent contexts. First is human activity detection from joint position information, and second is suspicious activity detection for tarmac security. It is shown that the first basis vector from factorization theorem is good enough to dierentiate activities for human data and to distinguish suspicious activities for tarmac security data. Second approach dierentiates individual components of those activities using semantic methodol- ogy. Although currently mainly used for improving search and information retrieval, we show that ontologies are applicable to video surveillance. We evaluate the domain ontologies from Challenge Project on Video Event Taxonomy sponsored by ARDA from the perspective of general ontology design principles. We also focused on the eect of the domain on the granularity of the ontology for suspicious activity detection.
  • Thumbnail Image
    Item
    Managing Uncertainty and Ontologies in Databases
    (2005-04-18) Hung, Edward; Subrahmanian, V.S.; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Nowadays a vast amount of data is generated in Extensible Markup Language (XML). However, it is necessary for applications in some domains to store and manipulate uncertain information, e.g. when the sensor inputs are noisy, or we want to store data that is uncertain. Another big change we can see in applications and web data is the increasing use of ontologies to describe the semantics of data, i.e., the semantic relationships between the terms in the databases. As such information is usually absent from traditional databases, there is tremendous opportunity to ask new kinds of queries that could not be handled in the past. This provides new challenges on how to manipulate and maintain such new kinds of database systems. In this dissertation, we will see how we can (i) incorporate and manipulate uncertainty in databases, and (ii) efficiently compute aggregates and maintain views on ontology databases. First, I explain applications that require manipulating uncertain information in XML databases and maintaining web ontology databases written in Resource Description Framework (RDF). I then introduce the probabilistic semistructured PXML data model with two formal semantics. I describe a set of algebraic operations and its efficient implementation. Aggregations of PXML instances are studied with two semantics proposed: possible-worlds semantics and expectation semantics. Efficient algorithms with pruning are given and evaluated to show their feasibility. I introduce PIXML, an interval probability version of PXML, and develop a formal semantics for it. A query language and its operational semantics are given and proved to be sound and complete. Based on XML, RDF is a language used to describe web ontologies. RDQL, an RDF query language, is extended to support view definition and aggregations. Two sets of algorithms are given to maintain non-aggregate and aggregate views. Experimental results show that they are efficient compared with standard relational view maintenance algorithms.