Physical Sciences and Mathematics Commons^™

Learning To Self-Train For Semi-Supervised Few-Shot Classification, Xinzhe Li, Qianru Sun, Yaoyao Liu, Shibao Zheng, Qin Zhou, Tat-Seng Chua, Bernt Schiele

Research Collection School Of Computing and Information Systems

Few-shot classification (FSC) is challenging due to the scarcity of labeled training data (e.g. only one labeled data point per class). Meta-learning has shown to achieve promising results by learning to initialize a classification model for FSC. In this paper we propose a novel semi-supervised meta-learning method called learning to self-train (LST) that leverages unlabeled data and specifically meta-learns how to cherry-pick and label such unsupervised data to further improve performance. To this end, we train the LST model through a large number of semi-supervised few-shot tasks. On each task, we train a few-shot model to predict pseudo labels for …

Growth Phase Proteomics Of The Heterotrophic Marine Bacterium Ruegeria Pomeroyi, Dasha Krayushkina, Emma Timmins-Schiffman, Jessica Faux, Damon H. May, Michael Riffle, H. Rodger Harvey, Brook L. Nunn

OES Faculty Publications

The heterotrophic marine bacterium, Ruegeria pomeroyi, was experimentally cultured under environmentally realistic carbon conditions and with a tracer-level addition of ¹³C-labeled leucine to track bacterial protein biosynthesis through growth phases. A combination of methods allowed observation of real-time bacterial protein production to understand metabolic priorities through the different growth phases. Over 2000 proteins were identified in each experimental culture from exponential and stationary growth phases. Within two hours of the ¹³C-labeled leucine addition, R. pomeroyi significantly assimilated the newly encountered substrate into new proteins. This dataset provides a fundamental baseline for understanding growth phase differences in molecular …

Overlap Matrix Completion For Predicting Drug-Associated Indications, Menhyun Yang, Huimin Luo, Yaohang Li, Fang-Xiang Wu, Jianxin Wang

Computer Science Faculty Publications

Identification of potential drug-associated indications is critical for either approved or novel drugs in drug repositioning. Current computational methods based on drug similarity and disease similarity have been developed to predict drug-disease associations. When more reliable drug- or disease-related information becomes available and is integrated, the prediction precision can be continuously improved. However, it is a challenging problem to effectively incorporate multiple types of prior information, representing different characteristics of drugs and diseases, to identify promising drug-disease associations. In this study, we propose an overlap matrix completion (OMC) for bilayer networks (OMC2) and tri-layer networks (OMC3) to predict potential drug-associated …

Comprehensive Investigation Of Bioactive Steroidal Alkaloids In Veratrum Californicum, Matthew West Turner

Boise State University Theses and Dissertations

Veratrum californicum, commonly referred to as corn lily or Californian false hellebore, grows in high mountain meadows and produces bioactive steroidal alkaloids, including cyclopamine. Cyclopamine is a potent inhibitor of the Hedgehog (Hh) signaling pathway. Our lab has optimized methods to extract cyclopamine and related steroidal alkaloids from V. californicum, and implemented in-house Hh signaling bioactivity evaluation of these steroidal alkaloids using a Shh-Light II cell assay. A survey of extraction methods and solvents resulted in the identification of conditions most favorable for alkaloid extraction yield and antagonist activity in the Shh-Light II cell assay. The highest amount …

Multilingual Information Retrieval: A Representation Building Perspective, Ion Madrazo

Boise State University Theses and Dissertations

Information Retrieval (IR) has changed the way we access digital resources and satisfy our daily information needs. Popular IR tools like Search Engines, Recommendation Systems, and Automatic Question Answering sites, act as a deterrent for information overload while fostering (at least in theory) the democratization of access to resources. Yet, in their majority, IR tools are built with a traditional user in mind. This causes users who deviate from the norm, e.g., users with low educational background, visually-impaired users, or users who speak different languages, to be undeserved and thus struggle to find the information they require. In this manuscript, …

The Spatial Distribution Of Elevated Uranium In The Treasure Valley Aquifer System, Southwest Idaho, Lloyd A. Womeldorph

Boise State University Theses and Dissertations

The Treasure Valley Aquifer System (TVAS) in southwestern Idaho contains well-documented uranium concentrations over the U.S. Environmental Protection Agency maximum contaminant level of 30 µg/L. With a population in the Treasure Valley projected to reach 1.6 million by 2065, in-depth horizontal and vertical spatial knowledge of the contaminant is needed. This study evaluates the horizontal and vertical spatial nature of uranium in the TVAS and interprets those observations to provide both a conceptual model of uranium behavior, and recommendations for water resource management. A large water quality dataset was compiled, and supplemented by data collected during a field sampling campaign, …

Detection And Countermeasure Of Saturation Attacks In Software-Defined Networks, Samer Yousef Khamaiseh

Boise State University Theses and Dissertations

The decoupling of control and data planes in software-defined networking (SDN) facilitates orchestrating the network traffic. However, SDN suffers from critical security issues, such as DoS saturation attacks on the data plane. These attacks can exhaust the SDN component resources, including the computational resources of the control plane, create a high packet loss rate and a long delay in delivering the OpenFlow messages due to the bandwidth consumption of the OpenFlow connection channel, and exhausting the buffer memory of the data plane.

Currently, most of the existing machine learning detection methods rely on a predefined time-window to start analyzing the …

Informing Field Management Decisions To Enhance Alfalfa Seed Production Using Remote Sensing, Thomas V. Van Der Weide

Boise State University Theses and Dissertations

The development rate of alfalfa seed crop depends on both environmental conditions and management decisions. Crop management decisions, such as determining when to release pollinators to optimize pollination, can be informed by the identification of plant development stages from remote sensing data. I first identify what electromagnetic wavelengths are sensitive to alfalfa plant development stages using hyperspectral data. A Random Forest regression is used to determine the best Vegetation Index (VI) to monitor how much of the plant is covered in flower. The results indicate that Blue, Green, and Near-Infrared are the important electromagnetic wavelengths for the VI. Imagery collected …

Bullynet: Unmasking Cyberbullies On Social Networks, Aparna Sankaran

Boise State University Theses and Dissertations

Social media has changed the way people communicate with each other, and consecutively affected people's ability to empathize in both positive and negative ways. One of the most harmful consequences of social media is the rise of cyberbullying, which tends to be more sinister than traditional bullying given that online records typically live on the internet for quite a long time and are hard to control. In this thesis, we present a three-phase algorithm, called BullyNet, for detecting cyberbullies on Twitter social network. We exploit bullying tendencies by proposing a robust method for constructing a cyberbullying signed network. BullyNet analyzes …

Geomorphology And Geochronology Of Sand Ramps Adjoining The Coral Pink Sand Dunes, Kane County, Utah, Kerri E. Spuller

Boise State University Theses and Dissertations

Projected climate change is expected to have widespread impacts on arid regions of the world. Inland, continental dune fields are sensitive to changes in climate and the surrounding environment and are an important source for understanding interactions between Earth’s surface, atmosphere, and biosphere in drylands. Reconstructing past periods of aeolian activity from dune fields can provide insights into how landscapes have responded to previous changes in climate. However, the nature of aeolian systems to recycle older deposits leads to a preservation bias towards only the youngest deposits. Targeting geomorphic settings that are conducive to preserving older records is critical to …

Characterizing Dryland Ecosystems Using Remote Sensing And Dynamic Global Vegetation Modeling, Abdolhamid Dashtiahangar

Boise State University Theses and Dissertations

Drylands include all terrestrial regions where the production of crops, forage, wood and other ecosystem services are limited by water. These ecosystems cover approximately 40% of the earth terrestrial surface and accommodate more than 2 billion people (Millennium Ecosystem Assessment, 2005). Moreover, the interannual variability of the global carbon budget is strongly regulated by vegetation dynamics in drylands. Understanding the dynamics of such ecosystems is significant for assessing the potential for and impacts of natural or anthropogenic disturbances and mitigation planning, and a necessary step toward enhancing the economic and social well-being of dryland communities in a sustainable manner (Global …

Falcon: Framework For Anomaly Detection In Industrial Control Systems, Subin Sapkota

Boise State University Theses and Dissertations

Industrial Control Systems (ICS) are used to control physical processes in the nation's critical infrastructures. They are composed of subsystems that control physical processes by analyzing the information received from the sensors. Based on the state of the process, the controller issues control commands to the actuators. These systems are utilized in a wide variety of operations such as water treatment plants, power, and manufacturing, etc. While the safety and security of these systems are of high concern, recent reports have shown an increase in targeted attacks that are aimed at manipulating the physical processes to cause catastrophic consequences. This …

Server Assignment With Time-Varying Workloads In Mobile Edge Computing, Quynh Vo

Graduate Doctoral Dissertations

Mobile Edge Computing (MEC) has emerged as a viable technology for mobile operators to push computing resources closer to the users so that requests can be served locally without long-haul crossing of the network core, thus improving network efficiency and user experience. In MEC, commodity servers are deployed in the edge to form a distributed network of mini datacenters. A consequential task is to partition the user cells into groups, each to be served by an edge server, to maximize the offloading to the edge. The conventional setting for this problem in the literature is: (1) assume that the interaction …

Seer: An Explainable Deep Learning Midi-Based Hybrid Song Recommender System, Khalil Damak, Olfa Nasraoui

Faculty and Staff Scholarship

State of the art music recommender systems mainly rely on either matrix factorization-based collaborative filtering approaches or deep learning architectures. Deep learning models usually use metadata for content-based filtering or predict the next user interaction by learning from temporal sequences of user actions. Despite advances in deep learning for song recommendation, none has taken advantage of the sequential nature of songs by learning sequence models that are based on content. Aside from the importance of prediction accuracy, other significant aspects are important, such as explainability and solving the cold start problem. In this work, we propose a hybrid deep learning …

Almost Homeomorphisms And Inscrutability, Michael Steven Andersen

Theses and Dissertations

“Homeomorphic'' is the standard equivalence relation in topology. To a topologist, spaces which are homeomorphic to each other aren't merely similar to each other, they are the same space. We study a class of functions which are homeomorphic at “most'' of the points of their domains and codomains, but which may fail to satisfy some of the properties required to be a homeomorphism at a “small'' portion of the points of these spaces. Such functions we call “almost homeomorphisms.'' One of the nice properties of almost homeomorphisms is the preservation of almost open sets, i.e. sets which are “close'' to …

An Equivalence Of Shape And Deck Groups; Further Classification Of Sharkovskii Groups, Tyler Willes Hills

Theses and Dissertations

In part one we show that for a compact, metric, locally path-connected topological space X, the shape group of X - as defined in Foundations of Shape Theory by Mardesic and Segal - is isomorphic to the inverse limit of discrete homotopy groups introduced by Conrad Plaut and Valera Berestovskii. We begin by providing the reader preliminary definitions of the fundamental group of a topological space, inverse systems and inverse limits, the Shape Category, discrete homotopy groups, and culminate by providing an isomorphism of the shape and deck groups for peano continua. In part two we develop work and provide …

Feasibility And Acceptability Of A Rural, Pragmatic, Telemedicine‐ Delivered Healthy Lifestyle Programme, John A. Batsis, Auden C. Mcclure, Aaron B. Weintraub, David F. Kotz, Sivan Rotenberg, Summer B. Cook, Diane Gilbert-Diamond, Kevin Curtis, Courtney J. Stevens, Diane Sette, Richard I. Rothstein

Dartmouth Scholarship

Background: The public health crisis of obesity leads to increasing morbidity that are even more profound in certain populations such as rural adults. Live, two‐way video‐conferencing is a modality that can potentially surmount geographic barriers and staffing shortages. Methods: Patients from the Dartmouth‐Hitchcock Weight and Wellness Center were recruited into a pragmatic, single‐arm, nonrandomized study of a remotely delivered 16‐week evidence‐based healthy lifestyle programme. Patients were provided hardware and appropriate software allowing for remote participation in all sessions, outside of the clinic setting. Our primary outcomes were feasibility and acceptability of the telemedicine intervention, as well as potential effectiveness on …

Leveraging Defects Life-Cycle For Labeling Defective Classes, Bailey R. Vandehei

Master's Theses

Data from software repositories are a very useful asset to building dierent kinds of

models and recommender systems aimed to support software developers. Specically,

the identication of likely defect-prone les (i.e., classes in Object-Oriented systems)

helps in prioritizing, testing, and analysis activities. This work focuses on automated

methods for labeling a class in a version as defective or not. The most used methods

for automated class labeling belong to the SZZ family and fail in various circum-

stances. Thus, recent studies suggest the use of aect version (AV) as provided by

developers and available in the issue tracker such as …

A Compositional Analysis Of Artificial And Terrestrial Analog Martian Regolith Simulants, Beverly Watson Kemmerer

Theses and Dissertations

Over the past 50 years, significant advances in the knowledge of the topography, climate, and geology of Mars have taken place. The global Martian regolith composition is highly basaltic, primarily composed of pyroxene, plagioclase, and olivine, a mixture of Fe oxides and Fe-Ti oxides, and some alteration minerals, i.e. sulfates, carbonates, and clays. The particle size distribution of Martian regolith ranges from 1 µm to 1000 µm, with average Martian soil grain sizes being 250 µm to 300 µm. These regolith properties, as well as geotechnical properties, all play a part in the outcome of certain in situ resource utilization …

Building Datasets From Publicly Accessible Social Media Images For Biometric Analysis, Giordano Roberto Benitez Torres

Theses and Dissertations

The world is developing at a rapid pace, and some of its advancement can be accredited to technological innovations that are affecting many aspects of society. A field within computer science that is increasingly reaching many industries is Biometrics, specifically the area face recognition. Researchers, scientists, and organizations actively try to improve the performance of tools and algorithms. Nevertheless, for it to occur, there is a need for high-quality datasets to test and develop new techniques. Never had humanity, in the course of the history of civilization, produced massive amounts of data as it currently does. Social media networks play …

Improving The Classification Of Tiny Images For Forensic Analysis, Roba Jafar Alharbi

Theses and Dissertations

Forensics can be defined as the approach that connects with and uses in governments and different organizations in order to detect any malicious activity. Digital forensics has become an essential approach to cyber investigation. Image forensics is one of the most beneficial ways that are used in digital forensics in order to help investigators in cybercrimes. Therefore, investigators can discover some new evidence besides what is already available on their systems when they use some digital forensics techniques. This thesis focuses on identifying an image based on its contents, especially tiny images. We investigated ways to improve the performance of …

Analysis Of Interfaces For The Nonlinear Degenerate Second Order Parabolic Equations Modeling Diffusion-Convection Processes, Lamees Kadhim Ali Alzaki

Theses and Dissertations

Dissertation pursues analysis of the short-time evolution of interfaces or free boundaries for the non-negative solutions of the nonlinear degenerate second order parabolic partial differential equation (PDE) ut = ( u m ) xx +b ( u γ ) x , x ∈ R,t > 0; m > 1, γ > 0,b ∈ R (1) modeling diffusion-convection processes arising in fluid or gas flow in a porous media, plasma physics, population dynamics in mathematical biology and other applications. Due to the implicit degeneration (m > 1), PDE (1) it possesses a property of the finite speed of propagation and develops interfaces or free boundaries …

Nonlocal Boundary Value Problems For Linear Hyperbolic Systems With Two Independent Variables, Afrah Almutairi

Theses and Dissertations

Nonlocal boundary value problems in a characteristic rectangle for second order linear hyperbolic systems are considered. There are established: (i) Unimprovable sufficient conditions for general boundary value problems to possess the Fredholm property; (ii) Optimal sufficient conditions of unique solvability of general boundary value problems; (iii) Effective sufficient conditions for doubly periodic problems to possess the Fredholm property; (iv) Unimprovable sufficient conditions of unique solvability of doubly periodic problems; (v) Effective sufficient conditions for boundary value problems of periodic type to possess the Fredholm property; (vi) Unimprovable sufficient conditions of unique solvability of boundary value problems of periodic type; (vii) …

Stability Analysis Of Neutral Functional Differential Equations Arising In Partial Element Equivalent Circuit Models, Howard Michael Allison

Theses and Dissertations

Neutral Functional Differential Equations (NFDEs) arise in the study of the Partial Element Equivalent Circuit (PEEC) model with time delays. We present sufficient conditions for asymptotic stability and global stability in the delays of the PEEC NFDE’s, using Lyapunov-Razumikhin function methods.. We develop, for the first time, a standard mixing-type nonlinearity for the PEEC NFDEs. Introducing time invariant and time varying nonlinear perturbation to the PEEC NFDEs, we develop sufficient conditions for stability of the nonlinear perturbed PEEC NFDEs and convergence of the nonlinear system to the original stable linear autonomous system. We also develop sufficient conditions for stability and …

Some Free Boundary Problems For The Nonlinear Degenerate Multidimensional Parabolic Equations Modeling Reaction-Diffusion Processes, Amna Abu Weden

Theses and Dissertations

This dissertation presents a full classification of the short-time behavior of the interfaces or free boundaries for the nonlinear second order degenerate multidimensional parabolic partial differential equation (PDE) ut −∆u m +buβ = 0, x ∈ R N ,0 < t < T (1) with m > 0, β > 0,b ∈ R, arising in various applications in fluid mechanics, filtration of oil or gas in a porous media, plasma physics, reaction-diffusion equations in chemical kinetics, population dynamics in mathematical biology etc. as a mathematical model of nonlinear diffusion phenomena in the presence of the absorption or release of energy. Cauchy problem with compactly supported and nonnegative initial function …

Computational Models For Biological Locomotion In Gels, Hashim Mohammed Alshehri

Theses and Dissertations

We investigated Low Reynold’s Number Locomotion in two-phase biological gels. The gel is composed of two materials: a viscous fluid solvent phase and a viscoelastic polymer network phase. A novel Two-phase Immersed Boundary Method (IBM) is developed to simulate the complicated interactions between an elastic boundary and a mixture of two fluids with very different physical properties. A further extension of the method is developed for the case where fluids satisfy partial-slip and free-slip conditions on the elastic boundary. Our major conclusions are summarized as following: (i) Our numerical scheme is proved to be robust and efficient. It can successfully …

Gut Content Analysis Of Benthic Foraging Fish Compared To Invertebrate Infauna Distributions In Dredged Vs. Undredged Habitats, Danielle Louise Juzwick

Theses and Dissertations

The Indian River Lagoon (IRL) has experienced a loss in seagrasses due to organic sediment accumulation and harmful algal blooms. Seagrasses are an essential nursery habitat for many fishes and benthic invertebrates. IRL fish known to utilize benthic habitat and feed on invertebrates include Atlantic croaker (Micropogonias undulates), juvenile Red Drum (Sciaenops ocellatus), and Mojarra (Eucinostomus spp.). The most abundant major infaunal taxa of the IRL are crustaceans, polychaetes, and mollusks. Environmental dredging in the IRL’s Turkey Creek is intended to improve water quality, remove harmful sediments and restore benthic habitat. Such dredging may impact benthic invertebrate populations and, indirectly, …

An Estimation Of Light Attenuation In A Shallow Estuary Using In-Situ Observations And Multispectral Satellite Imagery, Samin Taosif Aziz

Theses and Dissertations

A methodology is used to predict the underwater light field using data from the WorldView-3 (WV-3) multispectral satellite sensor. WV-3 multispectral imagery is analyzed to estimate reflectance above the water surface, followed by application of empirical algorithms at different water depth locations to estimate light attenuation and penetration depths in shallow water. This approach uses (a) full bit resolution WV-3 imagery, (b) top of atmosphere spectral radiance to estimate reflectance at the sensor and (c) atmospheric corrections to estimate water surface reflectance just above the water. After the water reflectance is obtained, it is used to correct for water surface …

A Case Study Of Stratospheric Ozone Transport To The Northern San Francisco Bay Area And Sacramento Valley During Cabots 2016, Jodie Clark, Sen Chiao

Faculty Research, Scholarly, and Creative Activity

The California Baseline Ozone Transport Study (CABOTS) was a major air quality study that collected ozone measurements aloft between mid-May and mid-August of 2016. Aircraft measurements, ground-based lidar measurements, and balloon-borne ozonesondes collected precise upper-air ozone measurements across the central and Southern California valley. Utilizing daily ozonesonde data from Bodega Bay, California, and Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2), reanalysis data for 25 July to 14 August 2016, three stratospheric intrusion events are identified over Northern California influencing air masses above Bodega Bay and Sacramento simultaneously. Calculated percent daily changes in afternoon ozonesonde observations indicate increasing …

Towards The Synthesis Of An Artificial Fluorescent Siderophore, Ashley Burns

Honors Theses

Siderophores are molecules capable of binding ferric ions with high affinity (Ka = 1052). This thesis outlines our attempts to synthesize an artificial siderophore, which incorporates a signaling group to help monitor the uptake of ferric ions. The artificial siderophore is a tripodal molecular receptor synthesized in a five-step procedure with a proposed final sixth step. Rhodamine 6G azide (95.0% yield) was formed from rhodamine 6G and the synthesis of 2-azidoethanamine was modified from the literature. The tripodal arm portion of the sensor was successfully synthesized by protecting the catechol of 2,3-dihydroxybenzaldehyde to form 3-formyl-1,2-phenylene dimethanesulfonate (98.8% yield). The reduction …