Physical Sciences and Mathematics Commons^™

Graph-Based Latent Embedding, Annotation And Representation Learning In Neural Networks For Semi-Supervised And Unsupervised Settings, Ismail Ozsel Kilinc

USF Tampa Graduate Theses and Dissertations

Machine learning has been immensely successful in supervised learning with outstanding examples in major industrial applications such as voice and image recognition. Following these developments, the most recent research has now begun to focus primarily on algorithms which can exploit very large sets of unlabeled examples to reduce the amount of manually labeled data required for existing models to perform well. In this dissertation, we propose graph-based latent embedding/annotation/representation learning techniques in neural networks tailored for semi-supervised and unsupervised learning problems. Specifically, we propose a novel regularization technique called Graph-based Activity Regularization (GAR) and a novel output layer modification called …

Context-Based Human Activity Recognition Using Multimodal Wearable Sensors, Pratool Bharti

USF Tampa Graduate Theses and Dissertations

In the past decade, Human Activity Recognition (HAR) has been an important part of the regular day to day life of many people. Activity recognition has wide applications in the field of health care, remote monitoring of elders, sports, biometric authentication, e-commerce and more. Each HAR application needs a unique approach to provide solutions driven by the context of the problem. In this dissertation, we are primarily discussing two application of HAR in different contexts. First, we design a novel approach for in-home, fine-grained activity recognition using multimodal wearable sensors on multiple body positions, along with very small Bluetooth beacons …

Multi-Step Tokenization Of Automated Clearing House Payment Transactions, Privin Alexander

USF Tampa Graduate Theses and Dissertations

Since its beginnings in 1974, the Automated Clearing House (ACH) network has grown into one of the largest, safest, and most efficient payment systems in the world. An ACH transaction is an electronic funds transfer between bank accounts using a batch processing system.

Currently, the ACH Network moves almost $43 trillion and 25 billion electronic financial transactions each year. With the increasing movement toward an electronic, interconnected and mobile infrastructure, it is critical that electronic payments work safely and efficiently for all users. ACH transactions carry sensitive data, such as a consumer's name, account number, tax identification number, account holder …

Constructing A Clinical Research Data Management System, Michael C. Quintero

USF Tampa Graduate Theses and Dissertations

Clinical study data is usually collected without knowing what kind of data is going to be collected in advance. In addition, all of the possible data points that can apply to a patient in any given clinical study is almost always a superset of the data points that are actually recorded for a given patient. As a result of this, clinical data resembles a set of sparse data with an evolving data schema. To help researchers at the Moffitt Cancer Center better manage clinical data, a tool was developed called GURU that uses the Entity Attribute Value model to handle …

Strong-Dism: A First Attempt To A Dynamically Typed Assembly Language (D-Tal), Ivory Hernandez

USF Tampa Graduate Theses and Dissertations

Dynamically Typed Assembly Language (D-TAL) is not only a lightweight and effective solution to the gap generated by the drop in security produced by the translation of high-level language instructions to low-level language instructions, but it considerably eases up the burden generated by the level of complexity required to implement typed assembly languages statically. Although there are tradeoffs between the static and dynamic approaches, focusing on a dynamic approach leads to simpler, easier to reason about, and more feasible ways to understand deployment of types over monomorphically-typed or untyped intermediate languages. On this occasion, DISM, a simple but powerful and …

Healthcare It In Skilled Nursing And Post-Acute Care Facilities: Reducing Hospital Admissions And Re-Admissions, Improving Reimbursement And Improving Clinical Operations, Scott L. Hopes

USF Tampa Graduate Theses and Dissertations

Health information technology (HIT), which includes electronic health record (EHR) systems and clinical data analytics, has become a major component of all health care delivery and care management. The adoption of HIT by physicians, hospitals, post-acute care organizations, pharmacies and other health care providers has been accepted as a necessary (and recently, a government required) step toward improved quality, care coordination and reduced costs: “Better coordination of care provides a path to improving communication, improving quality of care, and reducing unnecessary emergency room use and hospital readmissions. LTPAC providers play a critical role in achieving these goals” (HealthIT.gov, 2013).

Though …

Enhancing The Internet Of Things Architecture With Flow Semantics, Allen Ronald Deserranno

USF Tampa Graduate Theses and Dissertations

Internet of Things (‘IoT’) systems are complex, asynchronous solutions often comprised of various software and hardware components developed in isolation of each other. These components function with different degrees of reliability and performance over an inherently unreliable network, the Internet. Many IoT systems are developed within silos that do not provide the ability to communicate or be interoperable with other systems and platforms. Literature exists on how these systems should be designed, how they should interoperate, and how they could be improved, but practice does not always consult literature.

The work brings together a proposed reference architecture for the IoT …

Lung Ct Radiomics: An Overview Of Using Images As Data, Samuel Hunt Hawkins

USF Tampa Graduate Theses and Dissertations

Lung cancer is the leading cause of cancer-related death in the United States and worldwide. Early detection of lung cancer can help improve patient outcomes, and survival prediction can inform plans of treatment. By extracting quantitative features from computed tomography scans of lung cancer, predictive models can be built that can achieve both early detection and survival prediction. To build these predictive models, first a detected lung nodule is segmented, then image features are extracted, and finally a model can be built utilizing image features to make predictions. These predictions can help radiologists improve cancer care.

Building predictive models based …

Essays On Crowdfunding: Exploring The Funding And Post-Funding Phases And Outcomes, Onochie Fan-Osuala

USF Tampa Graduate Theses and Dissertations

In the recent years, crowdfunding (a phenomenon where individuals collectively contribute money to back different goals and projects through the internet) has been gaining a lot of attention especially for its socio-economic impact. This dissertation explores this phenomenon in three distinct but related essays. The first essay explores the nature and dynamics of backers’ contributions and uses the insights generated to develop a forecasting model that can predict crowdfunding campaign outcomes. The second essay investigates how creators’ crowdfunding campaign design decisions impact their funding and post-funding outcomes. Interestingly, the essay highlights that certain crowdfunding campaign design decisions have differential effects …

Computing The Rectilinear Crossing Number Of K, Soundarya Revoori

USF Tampa Graduate Theses and Dissertations

Rectilinear crossing number of a graph is the number of crossing edges in a drawing with all straight line edges. The problem of drawing an n-vertex complete graph such that its rectilinear crossing number is minimum is known to be an NP-Hard problem. In this thesis, we present a heuristic that attempts to achieve the theoretical lower bound value of the rectilinear crossing number of a n+1 vertex complete graph from that of n vertices. Our algorithm accepts an optimal or near-optimal rectilinear drawing of K_n graph as input and tries to place a new node such that …

Flexible And Feasible Support Measures For Mining Frequent Patterns In Large Labeled Graphs, Jinghan Meng

USF Tampa Graduate Theses and Dissertations

In recent years, the popularity of graph databases has grown rapidly. This paper focuses on single-graph as an effective model to represent information and its related graph mining techniques. In frequent pattern mining in a single-graph setting, there are two main problems: support measure and search scheme. In this paper, we propose a novel framework for constructing support measures that brings together existing minimum-image-based and overlap-graph-based support measures. Our framework is built on the concept of occurrence / instance hypergraphs. Based on that, we present two new support measures: minimum instance (MI) measure and minimum vertex cover (MVC) measure, that …

(Dis)Enchanted: (Re)Constructing Love And Creating Community In The, Shannon A. Suddeth

USF Tampa Graduate Theses and Dissertations

This thesis examines a queer fan community for the television show Once Upon a Time (OUAT) that utilizes the social networking site Tumblr as their primary base of fan activity. The Swan Queen fan community is comprised of individuals that collectively support and celebrate a non-canon romantic relationship between two of the female lead characters of the show rather than the canonic, heterocentric relationships that occur between the two women and their respective male love interests. I answer two research questions in this study: First, how are members of the Swan Queen fan community developing counter narratives of …

Estimation Of Human Poses Categories And Physical Object Properties From Motion Trajectories, Mona Fathollahi Ghezelghieh

USF Tampa Graduate Theses and Dissertations

Despite the impressive advancements in people detection and tracking, safety is still a key barrier to the deployment of autonomous vehicles in urban environments [1]. For example, in non-autonomous technology, there is an implicit communication between the people crossing the street and the driver to make sure they have communicated their intent to the driver. Therefore, it is crucial for the autonomous car to infer the future intent of the pedestrian quickly. We believe that human body orientation with respect to the camera can help the intelligent unit of the car to anticipate the future movement of the pedestrians. To …

Active Cleaning Of Label Noise Using Support Vector Machines, Rajmadhan Ekambaram

USF Tampa Graduate Theses and Dissertations

Large scale datasets collected using non-expert labelers are prone to labeling errors. Errors in the given labels or label noise affect the classifier performance, classifier complexity, class proportions, etc. It may be that a relatively small, but important class needs to have all its examples identified. Typical solutions to the label noise problem involve creating classifiers that are robust or tolerant to errors in the labels, or removing the suspected examples using machine learning algorithms. Finding the label noise examples through a manual review process is largely unexplored due to the cost and time factors involved. Nevertheless, we believe it …

Lost In The Crowd: Are Large Social Graphs Inherently Indistinguishable?, Subramanian Viswanathan Vadamalai

USF Tampa Graduate Theses and Dissertations

Real social graphs datasets are fundamental to understanding a variety of phenomena, such as epidemics, crowd management and political uprisings, yet releasing digital recordings of such datasets exposes the participants to privacy violations. A safer approach to making real social network topologies available is to anonymize them by modifying the graph structure enough as to decouple the node identity from its social ties, yet preserving the graph characteristics in aggregate. At scale, this approach comes with a significant challenge in computational complexity.

This thesis questions the need to structurally anonymize very large graphs. Intuitively, the larger the graph, the easier …

An Anthropological Study Of Security Operations Centers To Improve Operational Efficiency, Sathya Chandran Sundaramurthy

USF Tampa Graduate Theses and Dissertations

Security Operation Centers (SOCs) have become an integral component of business organizations all over the world. The concept of a SOC has existed for a few years now yet there is no systematic study documenting the occurrences of their operations. A lack of documented operational knowledge makes it a challenge for security researchers interested in improving operational efficiency through algorithms, tools, and processes.

SOC environments operate under a secrecy culture as a result of which researchers are not trusted by analysts and their managers. This lack of trust leads to only superficial information through methods such as interviews. Moreover, security …

Experimental Analysis On The Feasibility Of Voice Based Symmetric Key Generation For Embedded Devices, Surya Bharat Kamineni

USF Tampa Graduate Theses and Dissertations

In this thesis, we present results of an experimental study in order to generate a secure cryptographic key from the user’s voice which is to be shared between two mobile devices. We identified two security threats related to this problem, discussed the challenges to design the key generation/sharing mechanism, and proposed a new protocol based on bloom filters that overcomes the two main attacks by the intruder. One is when the attacker places its device in the close vicinity of the location where the user attempts to generate/share the key in order to derive the key from eavesdropping on communication …

Adaptive Region-Based Approaches For Cellular Segmentation Of Bright-Field Microscopy Images, Hady Ahmady Phoulady

USF Tampa Graduate Theses and Dissertations

Microscopy image processing is an emerging and quickly growing field in medical imaging research area. Recent advancements in technology including higher computation power, larger and cheaper storage modules, and more efficient and faster data acquisition devices such as whole-slide imaging scanners contributed to the recent microscopy image processing research advancement. Most of the methods in this research area either focus on automatically process images and make it easier for pathologists to direct their focus on the important regions in the image, or they aim to automate the whole job of experts including processing and classifying images or tissues that leads …

Semantic Description Of Activities In Videos, Fillipe Dias Moreira De Souza

USF Tampa Graduate Theses and Dissertations

Description of human activities in videos results not only in detection of actions and objects but also in identification of their active semantic relationships in the scene. Towards this broader goal, we present a combinatorial approach that assumes availability of algorithms for detecting and labeling objects and actions, albeit with some errors. Given these uncertain labels and detected objects, we link them into interpretative structures using domain knowledge encoded with concepts of Grenander’s general pattern theory. Here a semantic video description is built using basic units, termed generators, that represent labels of objects or actions. These generators have multiple out-bonds, …

Mitigation Of Insider Attacks For Data Security In Distributed Computing Environments, Santosh Aditham

USF Tampa Graduate Theses and Dissertations

In big data systems, the infrastructure is such that large amounts of data are hosted away from the users. Information security is a major challenge in such systems. From the customer’s perspective, one of the big risks in adopting big data systems is in trusting the service provider who designs and owns the infrastructure, with data security and privacy. However, big data frameworks typically focus on performance and the opportunity for including enhanced security measures is limited. In this dissertation, the problem of mitigating insider attacks is extensively investigated and several static and dynamic run-time techniques are developed. The proposed …

Scalable Unsupervised Learning With Game Theory, Alireza Chakeri

USF Tampa Graduate Theses and Dissertations

Recently dominant sets, a generalization of the notion of the maximal clique to edge-weighted graphs, have proven to be an effective tool for unsupervised learning and have found applications in different domains. Although, they were initially established using optimization and graph theory concepts, recent work has shown fascinating connections with evolutionary game theory, that leads to the clustering game framework. However, considering size of today's data sets, existing methods need to be modified in order to handle massive data. Hence, in this research work, first we address the limitations of the clustering game framework for large data sets theoretically. We …

Tradeoffs In Protocol Designs For Collaborative Authentication, Jacob Venne

USF Tampa Graduate Theses and Dissertations

Authentication is a crucial tool used in access control mechanisms to verify a user’s identity. Collaborative Authentication (co-authentication) is a newly proposed authentication scheme designed to improve on traditional token authentication. Co-authentication works by using multiple user devices as tokens to collaborate in a challenge and authenticate a user request on single device.

This thesis adds two contributions to the co-authentication project. First, a detailed survey of applications that are suitable for adopting co-authentication is presented. Second, an analysis of tradeoffs between varying protocol designs of co-authentication is performed to determine whether, and how, any designs are superior to other …

Forging Blockchains: Spatial Production And Political Economy Of Decentralized Cryptocurrency Code/Spaces, Joe Blankenship

USF Tampa Graduate Theses and Dissertations

Cryptocurrencies and blockchains are increasingly used, implemented and adapted for numerous purposes; people and businesses are integrating these technologies into their practices and strategies, creating new political economies and spaces in and of everyday life. This thesis seeks to develop a foundation of geographic theory for the study of spatial production within and surrounding blockchain technologies focusing on acute studies of Bitcoin as cryptocurrency, Ethereum as digital marketplace, and their conditions of possibility as decentralized autonomous organizations. Utilizing concepts from Henri Lefebvre's Production of Space, this thesis situates blockchain technologies within the wider discussion about the political economy of …

Initial Comparative Empirical Usability Testing For The Collaborative Authentication System, Kim Bursum

USF Tampa Graduate Theses and Dissertations

The Collaborative Authentication (co-authentication) system is an authentication system that relies on some or all members of a pre-registered set of secure hardware tokens being concurrently present to an authentication server at the moment of authentication. Previous researchers have compared various embodiments of the co-authentication system to each other including using Quick Response (QR) codes/cellphone cameras and Near Field Communication (NFC) between tokens. This thesis concerns the initial design and implementation of empirical comparative testing mechanisms between one embodiment of the co-authentication system and other commonly used authentication systems. One contribution is the simulated standard user ID and password login …