Computer Sciences Commons^™

Toward A Scientific Investigation Of Convolutional Neural Networks, Anh Tran

Honors Theses

This thesis does not assume the reader is familiar with artificial neural networks. However, to keep the thesis concise, it assumes the reader is familiar with the standard Machine Learning concepts of training set, validation set, and test set [1]. Their usage is intended to help ensure that the Machine Learning system can generalize its training from input examples used during its training to “similar” kinds of examples never used during its training.

The concept of a Convolutional Neural Network (CNN) is one of the most successful computational concepts today for solving image classification problems. However, CNNs are difficult and …

Efficient Algorithms For Analyzing Large Scale Network Dynamics: Centrality, Community And Predictability, Sima Das

Doctoral Dissertations

"Large scale networks are an indispensable part of our daily life; be it biological network, smart grids, academic collaboration networks, social networks, vehicular networks, or the networks as part of various smart environments, they are fast becoming ubiquitous. The successful realization of applications and services over them depend on efficient solution to their computational challenges that are compounded with network dynamics. The core challenges underlying large scale networks, for example: determining central (influential) nodes (and edges), interactions and contacts among nodes, are the basis behind the success of applications and services. Though at first glance these challenges seem to be …

Classification Of Basal Cell Carcinoma Using Telangiectatic Vessels And Machine Learning, Hemanth Yadav Aradhyula

Masters Theses

“Basal cell carcinoma (BCC) is one of the most common types of skin cancer in the United States. Early detection of BCC by noninvasive techniques can decrease delay in treatment and save cost. A recent study estimated that 5.4 million cases of non-melanocytic skin cancer (NMSC) occur each year in the US. BCC accounts for 50% of NMSC cases. Telangiectasia, which appears in most BCCs is an important feature for identification of BCC for an automatic diagnostic system. In this thesis, three methods for detection of telangiectasia present in dermoscopy lesion image (DI) were proposed. Detected telangiectasia in DI was …

The Viability Of Advantg Deterministic Method For Synthetic Radiography Generation, Andrew Albert Bingham

Masters Theses

"Time sensitive and high resolution image simulations are needed for synthetic radiography generation. The standard stochastic approach requires lengthy run times with poor statistics at higher resolutions. The investigation of the viability of a deterministic approach to synthetic radiography image generation was explored. The aim was to analyze a computational time decrease over the stochastic method. ADVANTG was compared to MCNP in multiple scenarios including a Benchtop CT prototype, to simulate high resolution radiography images. By using ADVANTG deterministic code to simulate radiography images the computational time was found to decrease over 10 times compared to the MCNP stochastic approach"--Abstract, …

Cyber-Physical Security Of A Chemical Plant, Prakash Rao Dunaka

Masters Theses

"The increasing number of cyber attacks on industries demands immediate attention for providing more secure mechanisms to safeguard industries and minimize risks. A supervisory control and data acquisition (SCADA) system employing the distributed networks of sensors and actuators that interact with the physical environment is vulnerable to attacks that target the interface between the cyber and physical subsystems. These cyber attacks are typically malicious actions that cause undesired results in the cyber physical world, for example, the Stuxnet attack that targeted Iran's nuclear centrifuges. An attack that hijacks the sensors in an attempt to provide false readings to the controller …

Personalizing Education With Algorithmic Course Selection, Tyler Morrow

Masters Theses

"The work presented in this thesis utilizes context-aware recommendation to facilitate personalized education and assist students in selecting courses (or in non-traditional curricula, topics or modules) that meet curricular requirements, leverage their skills and background, and are relevant to their interests. The original research contribution of this thesis is an algorithm that can generate a schedule of courses with consideration of a student's profile, minimization of cost, and complete adherence to institution requirements. The research problem at hand - a constrained optimization problem with potentially conflicting objectives - is solved by first identifying a minimal sets of courses a student …

Decodable Network Coding In Wireless Network, Junwei Su

Masters Theses

"Network coding is a network layer technique to improve transmission efficiency. Coding packets is especially beneficial in a wireless environment where the demand for radio spectrum is high. However, to fully realize the benefits of network coding two challenging issues that must be addressed are: (1) Guaranteeing separation of coded packets at the destination, and (2) Mitigating the extra coding/decoding delay. If the destination has all the needed packets to decode a coded packet, then separation failure can be averted. If the scheduling algorithm considers the arrival time of coding pairs, then the extra delay can be mitigated. In this …

Multi Stage Recovery From Large Scale Failure In Interdependent Networks, Maria Angelin John Bosco

Masters Theses

"Node and link failures that usually cause limited damage in a single network, may cascade into large scale disasters in the case of interdependent networks, due to the dependencies that exist between them. Recovery from such failures may require multiple stages or steps for complete restoration of connection or flow between them. When critical services are disrupted, the order in which the broken elements are repaired affects the earliest possible recovery time of vital services. In a flow network, one order of restoration may restore more flow at an earlier stage than another. The paper aims to model an efficient …

Multiple Security Domain Model Of A Vehicle In An Automated Vehicle System, Uday Ganesh Kanteti

Masters Theses

"This thesis focuses on the security of automated vehicle platoons. Specifically, it examines the vulnerabilities that occur via disruptions of the information flows among the different types of sensors, the communications network and the control unit in each vehicle of a platoon. Multiple security domain nondeducibility is employed to determine whether the system can detect attacks. The information flows among the various domains provide insights into the vulnerabilities that exist in the system by showing if an attacker’s actions cannot be deduced. If nondeducibility is found to be true, then an attacker can create an undetectable attack. Defeating nondeducibility requires …

Multiple Security Domain Nondeducibility Air Traffic Surveillance Systems, Anusha Thudimilla

Masters Theses

"Traditional security models partition the security universe into two distinct and completely separate worlds: high and low level. However, this partition is absolute and complete. The partition of security domains into high and low is too simplistic for more complex cyber-physical systems (CPS). Absolute divisions are conceptually clean, but they do not reflect the real world. Security partitions often overlap, frequently provide for the high level to have complete access to the low level, and are more complex than an impervious wall. The traditional models that handle situations where the security domains are complex or the threat space is ill …

A New Reinforcement Learning Algorithm With Fixed Exploration For Semi-Markov Decision Processes, Angelo Michael Encapera

Masters Theses

"Artificial intelligence or machine learning techniques are currently being widely applied for solving problems within the field of data analytics. This work presents and demonstrates the use of a new machine learning algorithm for solving semi-Markov decision processes (SMDPs). SMDPs are encountered in the domain of Reinforcement Learning to solve control problems in discrete-event systems. The new algorithm developed here is called iSMART, an acronym for imaging Semi-Markov Average Reward Technique. The algorithm uses a constant exploration rate, unlike its precursor R-SMART, which required exploration decay. The major difference between R-SMART and iSMART is that the latter uses, in addition …

A Bounded Actor-Critic Algorithm For Reinforcement Learning, Ryan Jacob Lawhead

Masters Theses

"This thesis presents a new actor-critic algorithm from the domain of reinforcement learning to solve Markov and semi-Markov decision processes (or problems) in the field of airline revenue management (ARM). The ARM problem is one of control optimization in which a decision-maker must accept or reject a customer based on a requested fare. This thesis focuses on the so-called single-leg version of the ARM problem, which can be cast as a semi-Markov decision process (SMDP). Large-scale Markov decision processes (MDPs) and SMDPs suffer from the curses of dimensionality and modeling, making it difficult to create the transition probability matrices (TPMs) …

Uface: Your Universal Password No One Can See, Nicholas Steven Hilbert

Masters Theses

"With the advantage of not having to memorize long passwords, facial authentication has become a topic of interest among researchers. However, since many users store images containing their face on social networking sites, a new challenge emerges in preventing attackers from impersonating these users by using these online photos. Another problem with most current facial authentication protocols is that they require an unencrypted image of each registered user's face to compare against. Moreover, they might require the user's device to execute computationally expensive multiparty protocols which presents a problem for mobile devices with limited processing power. Finally, these authentication protocols …

Siam Data Mining "Brings It" To Annual Meeting, Jeremy Kepner, Sanjukta Bhowmick, Aydın Buluç, Rajmonda Caceres, R. Jordan Crouser, Vijay Gadepally, Ben Miller, Jennifer Webster

Computer Science: Faculty Publications

The Data Mining Activity Group is one of SIAM's most vibrant and dynamic activity groups. To better share our enthusiasm for data mining with the broader SIAM community, our activity group organized six minisymposia at the 2016 Annual Meeting. These minisymposia included 48 talks organized by 11 SIAM members on - GraphBLAS (Aydın Buluç) - Algorithms and statistical methods for noisy network analysis (Sanjukta Bhowmick & Ben Miller) - Inferring networks from non-network data (Rajmonda Caceres, Ivan Brugere & Tanya Y. Berger-Wolf) - Visual analytics (Jordan Crouser) - Mining in graph data (Jennifer Webster, Mahantesh Halappanavar & Emilie Hogan) - …

New Principles For Auxetic Periodic Design, Ciprian Borcea, Ileana Streinu

Computer Science: Faculty Publications

We show that, for any given dimension d ≥ 2, the range of distinct possible designs for periodic frameworks with auxetic capabilities is infinite. We rely on a purely geometric approach to auxetic trajectories developed within our general theory of deformations of periodic frameworks.

End-To-End Molecular Communication Channels In Cell Metabolism: An Information Theoretic Study, Zahmeeth Sayed Sakkaff, Jennie L. Catlett, Mikaela Cashman, Massimiliano Pierobon, Nicole R. Buan, Myra B. Cohen, Christine A. Kelley

Department of Biochemistry: Faculty Publications

The opportunity to control and fine-tune the behavior of biological cells is a fascinating possibility for many diverse disciplines, ranging from medicine and ecology, to chemical industry and space exploration. While synthetic biology is providing novel tools to reprogram cell behavior from their genetic code, many challenges need to be solved before it can become a true engineering discipline, such as reliability, safety assurance, reproducibility and stability. This paper aims to understand the limits in the controllability of the behavior of a natural (non-engineered) biological cell. In particular, the focus is on cell metabolism, and its natural regulation mechanisms, and …

Abusive Text Detection Using Neural Networks, Hao Chen, Susan Mckeever, Sarah Jane Delany

Articles

Neurall network models have become increasingly popular for text classification in recent years. In particular, the emergence of word embeddings within deep learning architecture has recently attracted a high level of attention amongst researchers.

Data Protection And Humanitarian Emergencies, Fred H. Cate, Christopher Kuner, Dan Jerker B. Svantesson, Orla Lynskey, Christopher Millard

Articles by Maurer Faculty

No abstract provided.

The Benefits Of Task And Cognitive Workload Support For Operators In Ground Handling, Maria Chiara Leva, Yilmar Builes

Conference papers

The scope of the present work is to report an action research project applied to the relationship of task and cognitive workload support on one of the most important aspects of an airport: ground handling. At the beginning of the project workload management was not in the scope of work but as the project progressed and preliminary results and feedback were gained the researcher came to realize that some form of workload management support was also achieved as a by-product. The present paper is an attempt to account for what was achieved and how. Safe and efficient ground handling during …

Enhancing Existing Disaster Recovery Plans Using Backup Performance Indicators, Gwen White

Walden Dissertations and Doctoral Studies

Companies that perform data backup lose valuable data because they lack reliable data backup or restoration methods. The purpose of this study was to examine the need for a Six Sigma data backup performance indicator tool that clarifies the current state of a data backup method using an intuitive numerical scale. The theoretical framework for the study included backup theory, disaster recovery theory, and Six Sigma theory. The independent variables were implementation of data backup, data backup quality, and data backup confidence. The dependent variable was the need for a data backup performance indicator. An adapted survey instrument that measured …

Strategies To Prevent Security Breaches Caused By Mobile Devices, Tony Griffin

Walden Dissertations and Doctoral Studies

Data breaches happen almost every day in the United States and, according to research, the majority of these breaches occur due to a lack of security with organizations' mobile devices. Although most of the security policies related to mobile devices currently in place may meet the guidelines required by law, they often fail to prevent a data breach caused by a mobile device. The main purpose of this qualitative single case study was to explore the strategies used by security managers to prevent data breaches caused by mobile devices. The study population consisted of security managers working for a government …

Special Issue: Neutrosophic Theories Applied In Engineering, Florentin Smarandache, Jun Ye

Branch Mathematics and Statistics Faculty and Staff Publications

Neutrosophic sets and logic are generalizations of fuzzy and intuitionistic fuzzy sets and logic. Neutrosophic sets and logic are gaining significant attention in solving many real life decision making problems that involve uncertainty, impreciseness, vagueness, incompleteness, inconsistent, and indeterminacy. They have been applied in computational intelligence, multiple criteria decision making, image processing, medical diagnoses, etc. This Special Issue presents original research papers that report on state-of-the-art and recent advancements in neutrosophic sets and logic in soft computing, artificial intelligence, big and small data mining, decision making problems, and practical achievements.

Plithogeny, Plithogenic Set, Logic, Probability, And Statistics, Florentin Smarandache

Branch Mathematics and Statistics Faculty and Staff Publications

In this book we introduce for the first time, as generalization of dialectics and neutrosophy, the philosophical concept called plithogeny. And as its derivatives: the plithogenic set (as generalization of crisp, fuzzy, intuitionistic fuzzy, and neutrosophic sets), plithogenic logic (as generalization of classical, fuzzy, intuitionistic fuzzy, and neutrosophic logics), plithogenic probability (as generalization of classical, imprecise, and neutrosophic probabilities), and plithogenic statistics (as generalization of classical, and neutrosophic statistics).

Plithogeny is the genesis or origination, creation, formation, development, and evolution of new entities from dynamics and organic fusions of contradictory and/or neutrals and/or non-contradictory multiple old entities.

Plithogenic …

Crowdsourcing A Parallel Corpus For Conceptual Analysis Of Natural Language, Jamie C. Macbeth, Sandra Grandic

Computer Science: Faculty Publications

Computer users today are demanding greater performance from systems that understand and respond intelligently to human language as input. In the past, researchers proposed and built conceptual analysis systems that attempted to understand language in depth by decomposing a text into structures representing complex combinations of primitive acts, events, and state changes in the world the way people conceive them. However, these systems have traditionally been time-consuming and costly to build and maintain by hand. This paper presents two studies of crowdsourcing a parallel corpus to build conceptual analysis systems through machine learning. In the first study, we found that …

Fourth-Generation Fan Assessment Numeration System (Fans) Design And Performance Specifications, Michael P. Sama, George B. Day, Laura M. Pepple, Richard S. Gates

Biosystems and Agricultural Engineering Faculty Publications

The Fan Assessment Numeration System (FANS) is a measurement device for generating ventilation fan performance curves. Three different-sized FANS currently exist for assessing ventilation fans commonly used in poultry and livestock housing systems. All FANS consist of an array of anemometers inside an aluminum shroud that traverse the inlet or outlet of a ventilation fan. The FANS design has been updated several times since its inception and is currently in its fourth-generation (G4). The current design iteration (FANS-G4) is reported in this article with an emphasis on the hardware and software control, data acquisition systems, and operational reliability. Six FANS-G4 …

Novel Neuroevolution Techniques For The Life Science Domain, Timothy Manning

Theses

The life science domain is a high value research area, both in terms of the benefits in increased knowledge and in societal impact. Much of the research funding has focused on wet lab based approaches to increase visibility into biological processes and producing maximal relevant information on which to make decisions. Given the complexity of biological functions, in many cases this has led to an information overload. Researchers are now able to routinely generate and access petabytes of data as a result of high throughput experiments, and this capability is growing. This data can be difficult to interpret and intractable …

Diffusion Maps And Transfer Subspace Learning, Olga L. Mendoza-Schrock

Browse all Theses and Dissertations

Transfer Subspace Learning has recently gained popularity for its ability to perform cross-dataset and cross-domain object recognition. The ability to leverage existing data without the need for additional data collections is attractive for Aided Target Recognition applications. For Aided Target Recognition (or object assessment) applications, Transfer Subspace Learning is particularly useful, as it enables the incorporation of sparse and dynamically collected data into existing systems that utilize large databases. In this dissertation, Manifold Learning and Transfer Subspace Learning are combined to create new Aided Target Recognition systems capable of achieving high target recognition rates for cross-dataset conditions and cross-domain applications. …

Settings Protection Add-On: A User-Interactive Browser Extension To Prevent The Exploitation Of Preferences, Venkata Naga Siva Seelam

Browse all Theses and Dissertations

The abuse of browser preferences is a significant application security issue, despite numerous protections against automated software changing these preferences. Browser hijackers modify user’s desired preferences by injecting malicious software into the browser. Users are not aware of these modifications, and the unwanted changes can annoy the user and circumvent security preferences. Reverting these changes is not easy, and users often have to go through complicated sequences of steps to restore their preferences to the previous values. Tasks to resolve this issue include uninstalling and re-installing the browser, resetting browser preferences, and installing malware removal tools. This thesis describes a …

Exploiting Alignments In Linked Data For Compression And Query Answering, Amit Krishna Joshi

Browse all Theses and Dissertations

Linked data has experienced accelerated growth in recent years due to its interlinking ability across disparate sources, made possible via machine-processable RDF data. Today, a large number of organizations, including governments and news providers, publish data in RDF format, inviting developers to build useful applications through reuse and integration of structured data. This has led to tremendous increase in the amount of RDF data on the web. Although the growth of RDF data can be viewed as a positive sign for semantic web initiatives, it causes performance bottlenecks for RDF data management systems that store and provide access to data. …

A Work-Stealing For Dynamic Workload Balancing On Cpu-Gpu Heterogeneous Computing Platforms, Esraa A. Gad

Electronic Theses and Dissertations

Although many general purpose workloads have been accelerated on graphical processing units (gpus) over the last decade, other applications whose runtime behaviors are dynamic and irregular such as ones based on trees and graphs have suffered from serious workload imbalance problem caused by architectural differences between cpu and gpu processors. In this thesis, we propose a work-stealing framework to overcome such problems. Our proposed framework allows cpu and gpu threads to steal tasks from each other as well as within the same device by leveraging fine-grained data sharing and thread communication feature available on modern cpu-gpu heterogeneous systems. The implementation …