Physical Sciences and Mathematics Commons^™

Context-Driven Behavior: Improved Contextual Reasoning For Context-Aware Agents, Christian L. Wilson

Electronic Theses and Dissertations

Over the last three decades, a considerable amount of research has been dedicated to improving an artificial agent's ability to recognize and deal effectively with context. In this paper, I discuss a framework for a novel form of contextual reasoning. Unlike existing contextual reasoning frameworks, which allow an agent to apply its contextual knowledge after it is operating in an instance of a known context, the model I discuss allows an agent to reason about context proactively. With a proactive model, an agent forecasts the future contexts it will encounter, then takes steps to ensure its behaviors are appropriate for …

Accessible Autonomy: Exploring Inclusive Autonomous Vehicle Design And Interaction For People Who Are Blind And Visually Impaired, Paul D. S. Fink

Electronic Theses and Dissertations

Autonomous vehicles are poised to revolutionize independent travel for millions of people experiencing transportation-limiting visual impairments worldwide. However, the current trajectory of automotive technology is rife with roadblocks to accessible interaction and inclusion for this demographic. Inaccessible (visually dependent) interfaces and lack of information access throughout the trip are surmountable, yet nevertheless critical barriers to this potentially lifechanging technology. To address these challenges, the programmatic dissertation research presented here includes ten studies, three published papers, and three submitted papers in high impact outlets that together address accessibility across the complete trip of transportation. The first paper began with a thorough …

A Quantitative Visualization Tool For The Assessment Of Mammographic Risky Dense Tissue Types, Margaret R. Mccarthy

Electronic Theses and Dissertations

Breast cancer is the second most occurring cancer type and is ranked fifth in terms of mortality. X-ray mammography is the most common methodology of breast imaging and can show radiographic signs of cancer, such as masses and calcifcations. From these mammograms, radiologists can also assess breast density, which is a known cancer risk factor. However, since not all dense tissue is cancer-prone, we hypothesize that dense tissue can be segregated into healthy vs. risky subtypes. We propose that risky dense tissue is associated with tissue microenvironment disorganization, which can be quantified via a computational characterization of the whole breast …

Areas Of Same Cardinal Direction, Periyandy Thunendran

Electronic Theses and Dissertations

Cardinal directions, such as North, East, South, and West, are the foundation for qualitative spatial reasoning, a common field of GIS, Artificial Intelligence, and cognitive science. Such cardinal directions capture the relative spatial direction relation between a reference object and a target object, therefore, they are important search criteria in spatial databases. The projection-based model for such direction relations has been well investigated for point-like objects, yielding a relation algebra with strong inference power. The Direction Relation Matrix defines the simple region-to-region direction relations by approximating the reference object to a minimum bounding rectangle. Models that capture the direction between …

Artificial Dendritic Neuron: A Model Of Computation And Learning Algorithm, Zachary Hutchinson

Electronic Theses and Dissertations

Dendrites are root-like extensions from the neuron cell body and have long been thought to serve as the predominant input structures of neurons. Since the early twentieth century, neuroscience research has attempted to define the dendrite’s contribution to neural computation and signal integration. This body of experimental and modeling research strongly indicates that dendrites are not just input structures but are crucial to neural processing. Dendritic processing consists of both active and passive elements that utilize the spatial, electrical and connective properties of the dendritic tree.

This work presents a neuron model based around the structure and properties of dendrites. …

Performance Enhancement Of Hyperspectral Semantic Segmentation Leveraging Ensemble Networks, Nicholas Soucy

Electronic Theses and Dissertations

Hyperspectral image (HSI) semantic segmentation is a growing field within computer vision, machine learning, and forestry. Due to the separate nature of these communities, research applying deep learning techniques to ground-type semantic segmentation needs improvement, along with working to bring the research and expectations of these three communities together. Semantic segmentation consists of classifying individual pixels within the image based on the features present. Many issues need to be resolved in HSI semantic segmentation including data preprocessing, feature reduction, semantic segmentation techniques, and adversarial training. In this thesis, we tackle these challenges by employing ensemble methods for HSI semantic segmentation. …

Toward Improving Understanding Of The Structure And Biophysics Of Glycosaminoglycans, Elizabeth K. Whitmore

Electronic Theses and Dissertations

Glycosaminoglycans (GAGs) are the linear carbohydrate components of proteoglycans (PGs) that mediate PG bioactivities, including signal transduction, tissue morphogenesis, and matrix assembly. To understand GAG function, it is important to understand GAG structure and biophysics at atomic resolution. This is a challenge for existing experimental and computational methods because GAGs are heterogeneous, conformationally complex, and polydisperse, containing up to 200 monosaccharides. Molecular dynamics (MD) simulations come close to overcoming this challenge but are only feasible for short GAG polymers. To address this problem, we developed an algorithm that applies conformations from unbiased all-atom explicit-solvent MD simulations of short GAG polymers …

Approximating Fol Ontologies Using Owl2, Robert W. Powell

Electronic Theses and Dissertations

With the amount of data collected everyday ever expanding, techniques which allow com- puters to semantically understand data are growing in importance. Ontologies are a tool to describe the relationships connecting data so that computers can correctly interpret and combine data from many sources. An ontology about water needs to describe what the term "river" may refer to: An arbitrary river or one usable for navigation; a single tributary or an entire river network; the riverbed or the water itself? Well-designed ontologies can be shared, reused, and extended across multiple applications and facilitate betters integration of different data collections.

Common …

First-Year Computer Science Students: Pathways And Perceptions In Introductory Computer Science Courses, Christina A. Leblanc

Electronic Theses and Dissertations

This study examined student perceptions and experiences of an introductory Computer Science course at the University of Maine; COS 125: Introduction to Problem Solving Using Computer Programs. It also explored the pathways that students pursue after taking COS 125, depending on their success in the course, and their motivation to persist. Through characterizing student populations and their performance in their first semester in the Computer Science program, they can be placed into one of three categories that explain their path; a “continuer” (passed COS 125 and decided to stay in the major), a “persister” (did not pass COS 125 and …

A Qualitative Representation Of Spatial Scenes In R2 With Regions And Lines, Joshua Lewis

Electronic Theses and Dissertations

Regions and lines are common geographic abstractions for geographic objects. Collections of regions, lines, and other representations of spatial objects form a spatial scene, along with their relations. For instance, the states of Maine and New Hampshire can be represented by a pair of regions and related based on their topological properties. These two states are adjacent (i.e., they meet along their shared boundary), whereas Maine and Florida are not adjacent (i.e., they are disjoint).

A detailed model for qualitatively describing spatial scenes should capture the essential properties of a configuration such that a description of the represented objects …

A Longitudinal Study Of Mammograms Utilizing The Automated Wavelet Transform Modulus Maxima Method, Brian C. Toner

Electronic Theses and Dissertations

Breast cancer is a disease which predominatly affects women. About 1 in 8 women are diagnosed with breast cancer during their lifetime. Early detection is key to increasing the survival rate of breast cancer patients since the longer the tumor goes undetected, the more deadly it can become. The modern approach for diagnosing breast cancer relies on a combination of self-breast exams and mammography to detect the formation of tumors. However, this approach only accounts for tumors which are either detectable by touch or are large enough to be observed during a screening mammogram. For some individuals, by the time …

A Tale Of Two Bays: The Development And Applications Of The Saco And Casco Modeling Project, Stephen M. Moore

Electronic Theses and Dissertations

This thesis details the development and application of a finite-volume, hydrodynamic model of Saco and Casco Bays. The primary study conducted herein focused on coupling storm simulations with sea level rise (SLR) to identify vulnerabilities of the two bays. The February 1978 Northeaster and an April freshwater discharge event in 2007 following the Patriot’s Day Storm were modeled by utilizing the Finite-Volume Coastal Ocean Model (FVCOM). Both events were repeatedly simulated under SLR scenarios ranging from 0 to 7 ft. Modeled storm responses were identified from the 1978 blizzard simulations and were tracked across SLR scenarios. By comparing changes in …

Remote Sensing Of Icebergs In Greenland's Fjords And Coastal Waters, Jessica Scheick

Electronic Theses and Dissertations

Increases in ocean water temperature are implicated in driving recent accelerated rates of mass loss from the Greenland Ice Sheet. Icebergs provide a key tool for gaining insight into ice-ocean interactions and until recently have been relatively understudied. Here we develop several methods that exploit icebergs visible in optical satellite imagery to provide insight on the ice--ocean environment and explore how iceberg datasets can be used to examine the physics of iceberg decay and parent glacier properties. First, a semi-automated algorithm, which includes a machine learning-based cloud mask, is applied to six years (2000-2002 and 2013-2015) of the Landsat archive …

Three-Dimensional Bedrock Channel Evolution With Smoothed Particle Hydrodynamics, Nick Richmond

Electronic Theses and Dissertations

Bedrock channels are responsible for balancing and communicating tectonic and climatic signals across landscapes, but it is difficult and dangerous to observe and measure the flows responsible for removing weakly-attached blocks of bedrock from the channel boundary. Consequently, quantitative descriptions of the dynamics of bedrock removal are scarce. Detailed numerical simulation of violent flows in three dimensions has been historically challenging due to technological limitations, but advances in computational fluid dynamics aided by high-performance computing have made it practical to generate approximate solutions to the governing equations of fluid dynamics. From these numerical solutions we gain detailed knowledge of the …

Towards An Efficient, Scalable Stream Query Operator Framework For Representing And Analyzing Continuous Fields, John Whittier

Electronic Theses and Dissertations

Advancements in sensor technology have made it less expensive to deploy massive numbers of sensors to observe continuous geographic phenomena at high sample rates and stream live sensor observations. This fact has raised new challenges since sensor streams have pushed the limits of traditional geo-sensor data management technology. Data Stream Engines (DSEs) provide facilities for near real-time processing of streams, however, algorithms supporting representing and analyzing Spatio-Temporal (ST) phenomena are limited.

This dissertation investigates near real-time representation and analysis of continuous ST phenomena, observed by large numbers of mobile, asynchronously sampling sensors, using a DSE and proposes two novel stream …

Principles And Guidelines For Advancement Of Touchscreen-Based Non-Visual Access To 2d Spatial Information, Hari Prasath Palani

Electronic Theses and Dissertations

Graphical materials such as graphs and maps are often inaccessible to millions of blind and visually-impaired (BVI) people, which negatively impacts their educational prospects, ability to travel, and vocational opportunities. To address this longstanding issue, a three-phase research program was conducted that builds on and extends previous work establishing touchscreen-based haptic cuing as a viable alternative for conveying digital graphics to BVI users. Although promising, this approach poses unique challenges that can only be addressed by schematizing the underlying graphical information based on perceptual and spatio-cognitive characteristics pertinent to touchscreen-based haptic access. Towards this end, this dissertation empirically identified a …

Rediscovering The Interpersonal: Models Of Networked Communication In New Media Performance, Alicia Champlin

Electronic Theses and Dissertations

This paper examines the themes of human perception and participation within the contemporary paradigm and relates the hallmarks of the major paradigm shift which occurred in the mid-20th century from a structural view of the world to a systems view. In this context, the author’s creative practice is described, outlining a methodology for working with the communication networks and interpersonal feedback loops that help to define our relationships to each other and to media since that paradigm shift. This research is framed within a larger field of inquiry into the impact of contemporary New Media Art as we experience it. …

New Approaches To Mapping Forest Conditions And Landscape Change From Moderate Resolution Remote Sensing Data Across The Species-Rich And Structurally Diverse Atlantic Northern Forest Of Northeastern North America, Kasey R. Legaard

Electronic Theses and Dissertations

The sustainable management of forest landscapes requires an understanding of the functional relationships between management practices, changes in landscape conditions, and ecological response. This presents a substantial need of spatial information in support of both applied research and adaptive management. Satellite remote sensing has the potential to address much of this need, but forest conditions and patterns of change remain difficult to synthesize over large areas and long time periods. Compounding this problem is error in forest attribute maps and consequent uncertainty in subsequent analyses. The research described in this document is directed at these long-standing problems.

Chapter 1 demonstrates …

An Analysis Of International Agreements Over Cybersecurity, Lucas Ashbaugh

Electronic Theses and Dissertations

Research into the international agreements that increase cooperation over cybersecurity challenges is severely lacking. This is a necessary next step for bridging diplomatic challenges over cybersecurity. This work aspires to be push the bounds of research into these agreements and offer a tool that future researchers can rely on. For this research I created, and made publicly available, the International Cybersecurity Cooperation Dataset (ICCD), which contains over 350 international cybersecurity agreements and pertinent metadata. Each agreement is marked per which subtopics within cybersecurity related agreements it covers. These typologies are:

• Discussion and Dialogue

• Research

• Confidence Building Measures

• Incident Response

• Crime …

  ---

  Go to article

Using Hydroacoustics To Investigate Biological Responses In Fish Abundance To Restoration Efforts In The Penobscot River, Maine, Constantin C. Scherelis

Electronic Theses and Dissertations

Spatiotemporal advantages linked to hydroacoustic sampling techniques have caused a surge in the use of these techniques for fisheries monitoring studies applied over long periods of time in marine systems. Dynamic physical conditions such as tidal height, boat traffic, floating debris, and suspended particle concentrations result in unwanted noise signatures that vary in intensity and location within a hydroacoustic beam over time and can be mixed with the acoustic returns from intended targets (e.g., fish). Typical processing filters applied over long term datasets to minimize noise and maximize signals do not address spatiotemporal fluctuations of noise in dynamic systems. We …

Microstructural Analysis Of Thermoelastic Response, Nonlinear Creep, And Pervasive Cracking In Heterogeneous Materials, Alden C. Cook

Electronic Theses and Dissertations

This dissertation is concerned with the development of robust numerical solution procedures for the generalized micromechanical analysis of linear and nonlinear constitutive behavior in heterogeneous materials. Although the methods developed are applicable in many engineering, geological, and materials science fields, three main areas are explored in this work. First, a numerical methodology is presented for the thermomechanical analysis of heterogeneous materials with a special focus on real polycrystalline microstructures obtained using electron backscatter diffraction techniques. Asymptotic expansion homogenization and finite element analysis are employed for micromechanical analysis of polycrystalline materials. Effective thermoelastic properties of polycrystalline materials are determined and compared …

Investigating High Speed Localization Microscopy Through Experimental Methods, Data Processing Methods, And Applications Of Localization Microscopy To Biological Questions, Andrew J. Nelson

Electronic Theses and Dissertations

Fluorescence Photoactivation Localization Microscopy(FPALM) and other super resolution localization microscopy techniques can resolve structures with nanoscale resolution. Unlike techniques of electron microscopy, they are also compatible with live cell and live animal studies, making FPALM and related techniques ideal for answering questions about the dynamic nature of molecular biology in living systems. Many processes in biology occur on rapid sub second time scales requiring the imaging technique to be capable of resolving these processes not just with a high enough spatial resolution, but with an appropriate temporal resolution. To that end, this Dissertation in part investigates high speed FPALM as …

Improving The Performance Of Ice Sheet Modeling Through Embedded Simulation, Christopher G. Dufour

Electronic Theses and Dissertations

Understanding the impact of global climate change is a critical concern for society at large. One important piece of the climate puzzle is how large-scale ice sheets, such as those covering Greenland and Antarctica, respond to a warming climate. Given such ice sheets are under constant change, developing models that can accurately capture their dynamics represents a significant challenge to researchers. The problem, however, is properly capturing the dynamics of an ice sheet model requires a high model resolution and simulating these models is intractable even for state-of-the-art supercomputers.

This thesis presents a revolutionary approach to accurately capture ice sheet …

Email Similarity Matching And Automatic Reply Generation Using Statistical Topic Modeling And Machine Learning, Zachery L. Schiller

Electronic Theses and Dissertations

Responding to email is a time-consuming task that is a requirement for most professions. Many people find themselves answering the same questions over and over, repeatedly replying with answers they have written previously either in whole or in part. In this thesis, the Automatic Mail Reply (AMR) system is implemented to help with repeated email response creation. The system uses past email interactions and, through unsupervised statistical learning, attempts to recover relevant information to give to the user to assist in writing their reply.

Three statistical learning models, term frequency-inverse document frequency (tf-idf), Latent Semantic Analysis (LSA), and Latent Dirichlet …

Spatiotemporal Wireless Sensor Network Field Approximation With Multilayer Perceptron Artificial Neural Network Models, François Neville

Electronic Theses and Dissertations

As sensors become increasingly compact and dependable in natural environments, spatially-distributed heterogeneous sensor network systems steadily become more pervasive. However, any environmental monitoring system must account for potential data loss due to a variety of natural and technological causes. Modeling a natural spatial region can be problematic due to spatial nonstationarities in environmental variables, and as particular regions may be subject to specific influences at different spatial scales. Relationships between processes within these regions are often ephemeral, so models designed to represent them cannot remain static. Integrating temporal factors into this model engenders further complexity.

This dissertation evaluates the use …

Topographic Signatures Of Geodynamics, Samuel G. Roy

Electronic Theses and Dissertations

The surface of the Earth retains an imperfect memory of the diverse geodynamic, climatic, and surface transport processes that cooperatively drive the evolution of Earth. In this thesis I explore the potential of using topographic analysis and landscape evolution models to unlock past and/or present evidence for geodynamic activity. I explore the potential isolated effects of geodynamics on landscape evolution, particularly focusing on two byproducts of tectonic strain: rock displacement and damage. Field evidence supports a strong correlation between rock damage and erodibility, and a numerical sensitivity analysis supports the hypothesis that an order of magnitude weakening in rock, well …

Complex-Systems Approach To Simulating The Sea Urchin Ecology, Graham Andrew Morehead

Electronic Theses and Dissertations

Stocks of the native sea urchin (Strongylocentrotus droebachiensis) dropped dramatically during the peak of the urchin fishery in the early 1990’s and have not recovered. The current regulatory regime is based on analytic population models and two monolithic zones. Analytic models are insufficiently complex to capture many features that cause demise or survival of an urchin population. Scale, or granularity size, is too coarse. In contrast, a complex-systems-based model is able to capture these features. Presented here is a fine-scale simulation of a sea urchin fishery in the Gulf of Maine which behaves like a complex system, i.e. exhibits patchiness …

Contextual Analysis Of Large-Scale Biomedical Associations For The Elucidation And Prioritization Of Genes And Their Roles In Complex Disease, Jeremy J. Jay

Electronic Theses and Dissertations

Vast amounts of biomedical associations are easily accessible in public resources, spanning gene-disease associations, tissue-specific gene expression, gene function and pathway annotations, and many other data types. Despite this mass of data, information most relevant to the study of a particular disease remains loosely coupled and difficult to incorporate into ongoing research. Current public databases are difficult to navigate and do not interoperate well due to the plethora of interfaces and varying biomedical concept identifiers used. Because no coherent display of data within a specific problem domain is available, finding the latent relationships associated with a disease of interest is …

An Auv Simulator For Incorporating Physical Feedback, James Brawn Jr.

Electronic Theses and Dissertations

Traditional physics simulators use mathematical models to represent a realistic environment. Natural processes, however, are difficult to mimic accurately. We present a simulator that has the capability to alter its model based on actual physical measurements.

The simulator runs as a server to which remote clients can connect and assume control of entities within the virtual environment. The simulator then sends position updates to clients according to its model. Clients have the option of then correcting the data in these updates, sending feedback to the server. The server adjusts its model to accord with the corrections, allowing for a more …

The Extended Kalman-Consensus Filter, Andrew Pellett

Electronic Theses and Dissertations

An algorithm called the extended Kalman-Consensus filter is developed as an extension of the Kalman-Consensus filter to the non-linear case. The extended Kalman- Consensus filter is a technique for estimating the state of a non-linear process disturbed by noise using multiple observations from a distributed set of sense nodes. All sense nodes attempt to estimate the same state by determining how their observations affect that state, and by communicating with neighbor nodes. The algorithm is designed to be more accurate through measurement diversity, scalable to a large number of nodes, and robust against loss of nodes during operation. Simulations are …