Computer Sciences Commons^™

Dedicated Hardware For Machine/Deep Learning: Domain Specific Architectures, Angel Izael Solis

Open Access Theses & Dissertations

Artificial intelligence has come a very long way from being a mere spectacle on the silver screen in the 1920s [Hml18]. As artificial intelligence continues to evolve, and we begin to develop more sophisticated Artificial Neural Networks, the need for specialized and more efficient machines (less computational strain while maintaining the same performance results) becomes increasingly evident. Though these “new” techniques, such as Multilayer Perceptron’s, Convolutional Neural Networks and Recurrent Neural Networks, may seem as if they are on the cutting edge of technology, many of these ideas are over 60 years old! However, many of these earlier models, at …

Correlational Analysis Of The Relationship Among Mastery Experience, Self-Efficacy, And Project Success, Olakunle Taofeek Lemboye

Walden Dissertations and Doctoral Studies

Project managers are important to organizational performance and survival because of their role in managing, controlling, and steering organizational projects to success. Research has shown that project failures are globally pervasive due to the shortage of experienced and well-skilled project managers. The purpose of this descriptive correlational study was to improve the current understanding of the relationships among project managers' project management experience, self-efficacy, and project success, for which the research questions were focused on in addition to the role of project management experience on self-efficacy and project success. The theoretical framework was based on the social cognitive theory. This …

Strategies To Recover From Satellite Communication Failures, Charles Lomotey

Walden Dissertations and Doctoral Studies

In natural and manmade disasters, inadequate strategies to recover from satellite communication (SATCOM) failures can affect the ability of humanitarian organizations to provide timely assistance to the affected populations. This single case study explored strategies used by network administrators (NAs) to recover from SATCOM failures in humanitarian operations. The study population were NAs in Asia, the Middle East, Central Africa, East Africa, and West Africa. Data were collected from semistructured interviews with 9 NAs and an analysis of network statistics for their locations. The resource-based view was used as the conceptual framework for the study. Using inductive analysis, 3 themes …

Strategies For Cloud Services Adoption In Saudi Arabia, Wessam Hussein Abdulghani Mahmoud

Walden Dissertations and Doctoral Studies

The adoption rate of cloud computing is low among business organizations in Saudi Arabia, despite the cost-€saving benefits of using cloud services. The purpose of this multiple case study was to explore the strategies that information technology (IT) leaders in the manufacturing industry in Saudi Arabia used to adopt cloud computing to reduce IT costs. The target population of this study consisted of 5 IT leaders from 5 different manufacturing companies in Saudi Arabia who successfully adopted cloud computing in their companies to reduce IT costs. Rogers's diffusion of innovation theory was the conceptual framework for this research. Data collected …

Exploring Information Technology Return On Investment Reports For Planning, Budgeting, And Implementation, Constantine Bitwayiki

Walden Dissertations and Doctoral Studies

The failure rate of new government information technology (IT) projects in developing countries is high, with 35% classified as total failures and approximately 50% as partial failures. The population for this study was 10 senior managers of a public sector organization in Uganda selected because of high IT project successes achieved through leveraging IT return on investment (ROI) reports. The purpose of this qualitative single-case study was to explore the strategies Ugandan senior public sector officials used to leverage IT ROI reports during planning, budgeting, and implementation of IT projects to reduce failure rates. The conceptual frameworks were the strategic …

Using Labeling Theory As A Guide To Examine The Patterns, Characteristics, And Sanctions Given To Cybercrimes, Brian K. Payne, Brittany Hawkins, Chunsheng Xin

Sociology & Criminal Justice Faculty Publications

Over the past decade, reports of cybercrime have soared across the globe. Criminologists agree that the increase in cybercrime stems from technological advancements that have changed all facets of societal interactions. While it is agreed that technology has shaped cybercrime, there is less understanding about the dynamics of cybercrime. In particular, some researchers have explored whether these offenses are simply traditional types of crime that are now carried out through different strategies, while others have argued that cybercrimes are, in fact, new types of crime. This ambiguity potentially limits prevention and intervention strategies. In an effort to build our understanding …

Curtus: An Nlp Tool To Map Job Skills To Academic Courses, Daniel Rockwell

Theses and Dissertations

Many businesses are burdened with the need to train students for the job instead of finding them prepared for it. Few business leaders feel that colleges prepare students for future jobs from day one. It can be a challenge for colleges to determine if their curricula meet the industry needs. Mapping industry needs to academic courses can be advantageous to both parties as it will allow colleges to be aligned with the industry needs and accordingly satisfy those needs and will allow the industry to hire better prepared graduates. In an attempt to address this, a system prototype that uses …

Design And Evaluation Of A Wearable System For Facial Privacy, Scott Griffith

Theses and Dissertations

Through the increasingly common use of devices that provide ubiquitous sensor data such as wearables, mobile phones, and Internet-connected devices of the sort, privacy challenges are becoming even more significant. One major challenge that requires more focus is bystanders' privacy, as there are too few solutions that solve the issue. Of the solutions available, many of them do not give bystanders a choice in how their private data is used, Bystanders' privacy has become an afterthought when it comes to data capture in the forms of photographs, videos, voice recordings, etc. and continues to remain that way. This thesis provides …

A Simulation Model For Estimating Human Error Probability, Nitisha Reddy Boyapati

Theses and Dissertations

This report describes the system dynamics architecture of a simulation model which estimates human error probability for humans performing certain tasks in a given scenario. Human error probability is estimated as a function of the type of tasks performed and the number of performance shaping factors. In this work, the Standardized Plant Analysis Risk-Human (SPAR-H) reliability analysis method is utilized for estimating the probability of human error. The system dynamics simulation model captures the cause and effect relationships of the SPAR-H defined performance shaping factors that affect human error and uses them to assess the overall human error probability of …

Adapting Financial Technology Standards To Blockchain Platforms, Gabriel Bello

Theses and Dissertations

Traditional payment systems have standards designed to keep transaction data secure, but blockchain systems are not in scope for such security standards. We compare the Payment Application Data Security Standard’s (PA-DSS) applicability towards transaction-supported blockchain platforms to test the standard’s applicability. By highlighting the differences in implementation on traditional and decentralized transaction platforms, we critique and adapt the standards to fit the decentralized model. In two case studies, we analyze the QTUM and Ethereum blockchain platforms’ industry compliance, as their payment platforms support transactions equivalent to that of applications governed by the PA-DSS. We determine QTUM’s and Ethereum’s capabilities to …

Instance Segmentation And Object Detection In Road Scenes Using Inverse Perspective Mapping Of 3d Point Clouds And 2d Images, Chungyup Lee

Electronic Theses and Dissertations

The instance segmentation and object detection are important tasks in smart car applications. Recently, a variety of neural network-based approaches have been proposed. One of the challenges is that there are various scales of objects in a scene, and it requires the neural network to have a large receptive field to deal with the scale variations. In other words, the neural network must have deep architectures which slow down computation. In smart car applications, the accuracy of detection and segmentation of vehicle and pedestrian is hugely critical. Besides, 2D images do not have distance information but enough visual appearance. On …

Building Recommendation Systems, Orion Davis

Williams Honors College, Honors Research Projects

Recommendation systems are pieces of software that suggest new items to a user. There are many moving parts to these systems including data, the actual recommendation model, processing data and finally displaying data. This project explores the role each part plays in the overall system and how to develop a recommendation system for beer from scratch. This project highlights the algorithm behind the recommendations and a user facing Android application.

Equilibrium Structures And Thermal Fluctuations In Interacting Monolayers, Emmanuel Rivera

Williams Honors College, Honors Research Projects

Coherency strains appear in interacting atomic monolayers due to differing bond lengths, which can arise from different materials or geometries. Examples include extended monolayers interacting with a substrate and the interacting walls of a multi-walled carbon nanotube. These strains can induce various equilibrium configurations, which we will analyze by means of a phenomenological model that incorporates forces from bond stretching and bending within each layer and the weak van der Waals interactions connecting the separate layers. We vary the strengths of each interaction to explore their effects on equilibrium structures, and the specific case of a two-walled carbon nanotube is …

Absorption Calculator: A Cross-Platform Application For Portable Data Analysis, Annmarie Kolbl

Williams Honors College, Honors Research Projects

Traditional spectrometers are expensive and non-portable, making them inaccessible to the public. This application will be used in conjunction with spectrometer hardware developed by Erie Open Systems. The hardware itself is 3D printed and, in addition to being portable, enables data to be collected easily. The purpose of this project is to create a cross-platform application capable of reading the output from the spectrometer hardware, calculating the absorbance levels of the sample against the control, and recording the data in tables stored on the cloud. The end result will be an application that runs on iOS and Android, and is …

Gms - Guest Management System, Ethan Clark

Williams Honors College, Honors Research Projects

This paper examines the benefits of custom built as opposed to licensed software, using a guest management system as a comparison. The University of Akron employs a web application that allows residents to check guests into each residence hall on campus. In addition to this paper, a custom web application was built to contrast against the university's current system and to recognize the issues raised by its employee staff. This paper is written using the software development lifecycle as its main structure; the planning and analysis through the development process is discussed at length.

Basketball Charts, Kevin Lewis

Williams Honors College, Honors Research Projects

The purpose of this project was to develop an interactive web application with access to a self-updating database of basketball statistics. This data would then be used to allow users to generate informative visuals about specific sets of players. Obtaining statistics from the National Basketball Association (NBA) for the 2018-19 season was the original target goal. By utilizing an open source and community driven API, this goal was successfully achieved. With the data in place, the development of the chart building tool that was intended to be the primary functionality of the web application could begin. Highcharts was used as …

Islands Of Fitness Compact Genetic Algorithm For Rapid In-Flight Control Learning In A Flapping-Wing Micro Air Vehicle: A Search Space Reduction Approach, Kayleigh E. Duncan

Browse all Theses and Dissertations

On-going effective control of insect-scale Flapping-Wing Micro Air Vehicles could be significantly advantaged by active in-flight control adaptation. Previous work demonstrated that in simulated vehicles with wing membrane damage, in-flight recovery of effective vehicle attitude and vehicle position control precision via use of an in-flight adaptive learning oscillator was possible. Most recent approaches to this problem employ an island-of-fitness compact genetic algorithm (ICGA) for oscillator learning. The work presented provides the details of a domain specific search space reduction approach implemented with existing ICGA and its effect on the in-flight learning time. Further, it will be demonstrated that the proposed …

Towards Data And Model Confidentiality In Outsourced Machine Learning, Sagar Sharma

Browse all Theses and Dissertations

With massive data collections and needs for building powerful predictive models, data owners may choose to outsource storage and expensive machine learning computations to public cloud providers (Cloud). Data owners may choose cloud outsourcing due to the lack of in-house storage and computation resources or the expertise of building models. Similarly, users, who subscribe to specialized services such as movie streaming and social networking, voluntarily upload their data to the service providers' site for storage, analytics, and better services. The service provider, in turn, may also choose to benefit from ubiquitous cloud computing. However, outsourcing to a public cloud provider …

Abusive And Hate Speech Tweets Detection With Text Generation, Abhishek Nalamothu

Browse all Theses and Dissertations

According to a Pew Research study, 41% of Americans have personally experienced online harassment and two-thirds of Americans have witnessed harassment in 2017. Hence, online harassment detection is vital for securing and sustaining the popularity and viability of online social networks. Machine learning techniques play a crucial role in automatic harassment detection. One of the challenges of using supervised approaches is training data imbalance. Existing text generation techniques can help augment the training data, but they are still inadequate and ineffective. This research explores the role of domain-specific knowledge to complement the limited training data available for training a text …

Llvm-Ir Based Decompilation, Ilsoo Jeon

Browse all Theses and Dissertations

Decompilation is a process of transforming an executable program into a source-like high-level language code, which plays an important role in malware analysis, and vulnerability detection. In this thesis, we design and implement the middle end of a decompiler framework, focusing on Low Level Language properties reduction using the optimization techniques, propagation and elimination. An open-source software tool, dagger, is used to translate binary code to LLVM (Low Level Virtual Machine) Intermediate Representation code. We perform data flow analysis and control flow analysis on the LLVM format code to generate high-level code using a Functional Programming Langauge (FPL), Haskell. The …

Queue: A Mobile Application For Collaborative Music Playlists, Vlad Mirea

Williams Honors College, Honors Research Projects

This paper focuses on the design and development of the mobile application “Queue”. Queue is an app for creating music playlists that anyone can add songs to while a host controls playback. The app connects to music streaming services such as Spotify to allow users access to their favorite songs while providing functionality not found in those services.

Microarray Data Analysis And Classification Of Cancers, Grant Gates

Williams Honors College, Honors Research Projects

When it comes to cancer, there is no standardized approach for identifying new cancer classes nor is there a standardized approach for assigning cancer tumors to existing classes. These two ideas are known as class discovery and class prediction. For a cancer patient to receive proper treatment, it is important that the type of cancer be accurately identified. For my Senior Honors Project, I would like to use this opportunity to research a topic in bioinformatics. Bioinformatics incorporates a few different subjects into one including biology, computer science and statistics. An intricate method for class discovery and class prediction is …

Web Service Composition Optimization, Hussain Aljafer

Wayne State University Dissertations

In recent years, users and organizations started switching from workstation- based applications to Web services also known as cloud services. Web services offer many advantages such as cost of use and maintenance. Web services follow the pay- per-use pricing model so the users pay for their usage only. Due to the huge number of services, a composition optimization mechanism is needed to help the users find the best service/set of services for their application/s. On the other hand, service providers look for generating the highest profit possible from the offered services. In this dissertation, we address the problem from both …

Toward Energy Efficient Systems Design For Data Centers, Bing Luo

Wayne State University Dissertations

Surge growth of numerous cloud services, Internet of Things, and edge computing promotes continuous increasing demand for data centers worldwide. Significant electricity consumption of data centers has tremendous implications on both operating and capital expense. The power infrastructure, along with the cooling system cost a multi-million or even billion dollar project to add new data center capacities. Given the high cost of large-scale data centers, it is important to fully utilize the capacity of data centers to reduce the Total Cost of Ownership. The data center is designed with a space budget and power budget. With the adoption of high-density …

Bundle: Taming The Cache And Improving Schedulability Of Multi-Threaded Hard Real-Time Systems, Corey Tessler

Wayne State University Dissertations

For hard real-time systems, schedulability of a task set is paramount. If a task set is not deemed schedulable under all conditions, the system may fail during operation and cannot be deployed in a high risk environment. Schedulability testing has typically been separated from worst-case execution time (WCET) analysis. Each task’s WCET value is calculated independently and provided as input to a schedulability test. However, a task’s WCET value is influenced by scheduling decisions and the impact of cache memory. Thus, schedulability tests have been augmented to include cache-related preemption delay (CRPD). From this classical perspective, the effect of cache …

Attention-Based Models For Deep Reinforcement Learning, Elaheh Barati

Wayne State University Dissertations

Attention mechanism has shown promising results in many fields of machine learning such as image captioning and machine translation. In this work, we focus on attention-based models for deep reinforcement learning. We concentrate on developing deep neural networks

that are fed with a sequence of high-dimensional raw pixels. Particularly, we design attention-based models for challenging tasks including navigation, autonomous driving, and video captioning. In these tasks, deep reinforcement learning algorithms facilitate training of their sophisticated models, and the attention mechanism serves different purposes. In the navigation and autonomous driving tasks, through the attention mechanism, our model attends over different views …

Utilizing Knowledge Bases In Information Retrieval For Clinical Decision Support And Precision Medicine, Saeid Balaneshinkordan

Wayne State University Dissertations

Accurately answering queries that describe a clinical case and aim at finding articles in a collection of medical literature requires utilizing knowledge bases in capturing many explicit and latent aspects of such queries. Proper representation of these aspects needs knowledge-based query understanding methods that identify the most important query concepts as well as knowledge-based query reformulation methods that add new concepts to a query. In the tasks of Clinical Decision Support (CDS) and Precision Medicine (PM), the query and collection documents may have a complex structure with different components, such as disease and genetic variants that should be transformed to …

The Use Of Cultural Algorithms To Learn The Impact Of Climate On Local Fishing Behavior In Cerro Azul, Peru, Khalid Kattan

Wayne State University Dissertations

Recently it has been found that the earth’s oceans are warming at a pace that is 40% faster than predicted by a United Nations panel a few years ago. As a result, 2019 has become the warmest year on record for the earth’s oceans. That is because the oceans have acted as a buffer by absorbing 93% of the heat produced by the greenhouse gases [40].

The impact of the oceanic warming has already been felt in terms of the periodic warming of the Pacific Ocean as an effect of the ENSO process. The ENSO process is a cycle of …

Learning From Heterogeneous Data, Lu Wang

Wayne State University Dissertations

Data with both heterogeneity and homogeneity is now ubiquitous due to the development of multitudinous data collection techniques. To encode the data heterogeneity and homogeneity, we focus on unsupervised and supervised learning approaches. In unsupervised learning, to consider both data heterogeneity and homogeneity, we develop three clustering frameworks to maximize the heterogeneity among data sub-groups and homogeneity within each data sub-group for over-dispersed data in three different data types, i.e., alphabetic, network and mixed feature types data. In supervised learning, the traditional approaches, however, either build a global model for a whole group including all sub-groups, which fail to consider …

3d Surface Registration Using Geometric Spectrum Of Shapes, Hajar Hamidian

Wayne State University Dissertations

Morphometric analysis of 3D surface objects are very important in many biomedical applications and clinical diagnoses. Its critical step lies in shape comparison and registration. Considering that the deformations of most organs such as heart or brain structures are non-isometric, it is very difficult to find the correspondence between the shapes before and after deformation, and therefore, very challenging for diagnosis purposes.

To solve these challenges, we propose two spectral based methods. The first method employs the variation of the eigenvalues of the Laplace-Beltrami operator of the shape and optimize a quadratic equation in order to minimize the distance between …