Engineering Commons^™

Evaluating Creative Choice In K-12 Computer Science Curriculum, Kirsten L. Mork

Master's Theses

Computer Science is an increasingly important topic in K-12 education. Ever since the "computing crisis" of the early 2000s, where enrollment in CS dropped by over half in a five year span, increasing research has gone into improving and broadening enrollment in CS courses. Research shows the importance of introducing CS at a young age and the need for more exposure for younger children and young adults alike in order to work towards equity in the field. While there are many reasons for disinterest in CS courses, studies found one reason young adults do not want to study CS is …

Long Term Software Quality And Reliability Assurance In A Small Company, Eric Abuta

Computer Science and Engineering Theses and Dissertations

Demonstrating software reliability across multiple software releases has become essential in making informed decisions of upgrading software releases without impacting significantly end users' characterized processes and software quality standards. Standard defect and workload data normally collected in a typical small software development organization can be used for this purpose. Objective of this study was to demonstrate how to measure software reliability in multiple releases and whether continuous defect fixes and code upgrades increased software reliability. This study looked at techniques such as trend test that evaluated software system's overall trend and stability, input domain reliability models (IDRM) that assessed system's …

Different Approaches To Blurring Digital Images And Their Effect On Facial Detection, Erich-Matthew Pulfer

Computer Science and Computer Engineering Undergraduate Honors Theses

The purpose of this thesis is to analyze the usage of multiple image blurring techniques and determine their effectiveness in combatting facial detection algorithms. This type of analysis is anticipated to reveal potential flaws in the privacy expected from blurring images or, rather, portions of images. Three different blurring algorithms were designed and implemented: a box blurring method, a Gaussian blurring method, and a differential privacy-based pixilation method. Datasets of images were collected from multiple sources, including the AT&T Database of Faces. Each of these three methods were implemented via their own original method, but, because of how common they …

Granny Pod Virtual Assistant, David Connolly, Bing Chen

Theses/Capstones/Creative Projects

Dr. Chen is working on a sustainable small house (SSH) project, sometimes called the “Granny Pod”. Regulations will soon allow homeowners to house their parents on their property, which can be an opportunity live independently in a cheap, sustainable, and convenient alternative to a retirement community. To help achieve this vision, a Virtual Assistant system for the SSH was developed. The system uses a Google Home or Amazon Echo to respond to the voice command “Hey Google (or Alexa), I need help” by contacting the nearby homeowner or caretaker. It alerts the resident who is at the door when the …

Application-Specific Memory Subsystem Benchmarking, Mahesh Lakshminarasimhan

Boise State University Theses and Dissertations

Application performance often depends on achieved memory bandwidth. Achieved memory bandwidth varies greatly given specific combinations of instruction mix and order, working set size, and access pattern. Achieving good application performance depends on optimizing these characteristics within the constraints of the given application. This task is complicated due to the lack of information about the impact of small changes on the performance. Some information is provided by benchmarks, but most memory benchmarks are confined to simple access patterns that are not representative of patterns found in real applications.

This thesis presents AdaptMemBench, a configurable benchmark framework designed to explore the …

Unicorn Framework: A User-Centric Approach Toward Formal Verification Of Privacy Norms, Rezvan Joshaghani

Boise State University Theses and Dissertations

In the development of complex systems, such as user-centric privacy management systems with multiple components and attributes, it is important to formalize the process and develop mathematical models that can be utilized to automatically make decisions on the information sharing actions of users. While valuable, the current state-of-the-art models are mostly based on enterprise/organizational privacy perspectives and leave the main actor, i.e., the user, uninvolved or with limited ability to control information sharing actions. These approaches cannot be applied to a user-centric environment since user privacy policies are dynamic because they change based on the information sharing context and environment. …

A Purely Defeasible Argumentation Framework, Zimi Li

Dissertations, Theses, and Capstone Projects

Argumentation theory is concerned with the way that intelligent agents discuss whether some statement holds. It is a claim-based theory that is widely used in many areas, such as law, linguistics and computer science. In the past few years, formal argumentation frameworks have been heavily studied and applications have been proposed in fields such as natural language processing, the semantic web and multi-agent systems. Studying argumentation provides results which help in developing tools and applications in these areas. Argumentation is interesting as a logic-based approach to deal with inconsistent information. Arguments are constructed using a process like logical inference, with …

Public Blockchain Scalability: Advancements, Challenges And The Future, Amritraj .

Master of Science in Software Engineering Theses

In the last decade, blockchain has emerged as one of the most influential innovations in software architecture and technology. Ideally, blockchains are designed to be architecturally and politically decentralized, similar to the Internet. But recently, public and permissionless blockchains such as Bitcoin and Ethereum have faced stumbling blocks in the form of scalability. Both Bitcoin and Ethereum process fewer than 20 transactions per second, which is significantly lower than their centralized counterpart such as VISA that can process approximately 1,700 transactions per second. In realizing this hindrance in the wide range adoption of blockchains for building advanced and large scalable …

A Grammar Based Approach To Distributed Systems Fault Diagnosis Using Log Files, Stephen Hanka

Computer Science and Engineering Theses and Dissertations

Diagnosing and correcting failures in complex, distributed systems is difficult. In a network of perhaps dozens of nodes, each of which is executing dozens of interacting applications, sometimes from different suppliers or vendors, finding the source of a system failure is a confusing, tedious piece of detective work. The person assigned this task must trace the failing command, event, or operation through the network components and find a deviation from the correct, desired interaction sequence. After a deviation is identified, the failing applications must be found, and the fault or faults traced to the incorrect source code.

Often the primary …

Bridgr: An Ios Application For Organizing And Discussing Long-Distance Carpooling, Harrison Engoren, Erik Zorn

Senior Theses

Bridgr is an iOS application that facilitates long distance carpooling. This application allows drivers to post destinations on an interactable map so that they can be linked with students that need a ride to a location within close proximity of the posted destination. The riders and driver are linked in a common chat board where they can discuss ride details among themselves. The goal of Bridgr is to allow drivers to utilize extra space in their car in turn for fellowship and/or gas money.

Distributed Wireless Algorithms For Rfid Systems: Grouping Proofs And Cardinality Estimation, Vanya D. Cherneva

LSU Doctoral Dissertations

The breadth and depth of the use of Radio Frequency Identification (RFID) are becoming more substantial. RFID is a technology useful for identifying unique items through radio waves. We design algorithms on RFID-based systems for the Grouping Proof and Cardinality Estimation problems.

A grouping-proof protocol is evidence that a reader simultaneously scanned the RFID tags in a group. In many practical scenarios, grouping-proofs greatly expand the potential of RFID-based systems such as supply chain applications, simultaneous scanning of multiple forms of IDs in banks or airports, and government paperwork. The design of RFID grouping-proofs that provide optimal security, privacy, and …

Machine Translation With Image Context From Mandarin Chinese To English, Brooke E. Johnson

Theses and Dissertations

Despite ongoing improvements in machine translation, machine translators still lack the capability of incorporating context from which source text may have been derived. Machine translators use text from a source language to translate it into a target language without observing any visual context. This work aims to produce a neural machine translation model that is capable of accepting both text and image context as a multimodal translator from Mandarin Chinese to English. The model was trained on a small multimodal dataset of 700 images and sentences, and compared to a translator trained only on the text associated with those images. …

Hyper-Parameter Optimization Of A Convolutional Neural Network, Steven H. Chon

Theses and Dissertations

In the world of machine learning, neural networks have become a powerful pattern recognition technique that gives a user the ability to interpret high-dimensional data whereas conventional methods, such as logistic regression, would fail. There exists many different types of neural networks, each containing its own set of hyper-parameters that are dependent on the type of analysis required, but the focus of this paper will be on the hyper-parameters of convolutional neural networks. Convolutional neural networks are commonly used for classifications of visual imagery. For example, if you were to build a network for the purpose of predicting a specific …

My Baseball Collection App, Nicolas A. Parra

Computer Science and Software Engineering

My Baseball Collection is an iOS application that aims to simplify the management and expansion of physical baseball trading card collections. The app allows users to digitize their baseball card collection by uploading images of cards they possess, creating a wishlist of cards they are seeking, and viewing the collections and wishlists of other users. This project seeks to provide quality of life improvements to those within the baseball card trading community and to further facilitate communication and trading in an online world.

Genet-Cnv: Boolean Implication Networks For Modeling Genome-Wide Co-Occurrence Of Dna Copy Number Variations, Salvi Singh

Graduate Theses, Dissertations, and Problem Reports

Lung cancer is the leading cause of cancer-related death in the world. Lung cancer can be categorized as non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). NSCLC makes up about 80% to 85% of lung cancer cases diagnosed, whereas SCLC is responsible for 10% to 15% of the cases. It remains a challenge for physicians to identify patients who shall benefit from chemotherapy. In such a scenario, identifying genes that can facilitate therapeutic target discoveries and better understanding disease mechanisms and their regulation in different stages of lung cancer, remains an important topic of research.

In this …

Security Bug Report Classification Using Feature Selection, Clustering, And Deep Learning, Tanner D. Gantzer

Graduate Theses, Dissertations, and Problem Reports

As the numbers of software vulnerabilities and cybersecurity threats increase, it is becoming more difficult and time consuming to classify bug reports manually. This thesis is focused on exploring techniques that have potential to improve the performance of automated classification of software bug reports as security or non-security related. Using supervised learning, feature selection was used to engineer new feature vectors to be used in machine learning. Feature selection changes the vocabulary used by selecting words with the greatest impact on classification. Feature selection was able to increase the F-Score across the datasets by increasing the precision. We also explored …

Integrating Multi-Source Weather Data For Deep Learning, Haidar A. Alanbari Mr

Dissertations and Theses

Big Data has been playing a major role in the domain of Deep Learning applications as many companies and institutions continue to find solutions and extract certain trends in fields of climate change, weather forecasting and meteorology. This project extracts weather events data from multiple data sources that are supported by National Centers for Environmental information (NCEI) [1] and Amazon Web Services (AWS) [2]. Data sources include Next-Generation NEXRAD [3] Doppler radar reflectivity, GOES-16 [4] multi-channel satellite imagery and NCEI [1] storm events. Then, it integrates and refines data in proper formats to be fed to the open-source Detectron [5] …

Geometric Correction For A Spherical Mirror Projection On A Nonplanar Surface, Methuen J. Bell-Isaac

Senior Projects Spring 2019

This paper discusses an approach for removing distortion from an image projected on a non-planar surface. With a single projector setup in a spherical mirror projection system, it becomes possible to preserve image features. The approach takes advantage of the configuration of the surface, specifically, the geodesic dome in this project. The configuration acts as a mold so that a warp mesh can be designed to match the surface configuration. Points in an image are then mapped to their corresponding point on the destination multi-planar surface represented by the mesh. The removal of distortion brings us a step closer to …

Car Image Classification Using Deep Neural Networks, Mingchen Li

Honors Theses

Image classification is widely used in many fields of study. Deep neural networks are proven to be effective classifier structure due to its massive parameters and training capability. This paper outlines the development of Deep Neural Network in recent years and applied them on a Car image data set in order to compare their performances.

The Evaluation Of An Android Permission Management System Based On Crowdsourcing, Pulkit Rustgi

Theses and Dissertations

Mobile and web application security, particularly concerning the area of data privacy, has received much attention from the public in recent years. Most applications are installed without disclosing full information to users and clearly stating what they have access to. This often raises concerns when users become aware of unnecessary information being collected or stored. Unfortunately, most users have little to no technical knowledge in regard to what permissions should be granted and can only rely on their intuition and past experiences to make relatively uninformed decisions. DroidNet, a crowdsource based Android recommendation tool and framework, is a proposed avenue …

Investigation And Development Of Exhaust Flow Rate Estimation Methodologies For Heavy-Duty Vehicles, Chakradhar Reddy Vardhireddy

Graduate Theses, Dissertations, and Problem Reports

Exhaust gas flow rate from a vehicle tailpipe has a great influence on emission mass rate calculations, as the emission fractions of individual gases in the exhaust are calculated by using the measured exhaust flow rate. The development of high-end sensor technologies and emission pollutant measurement instruments, which can give instantaneous values of volume concentration of pollutants flowing out of the engine are gaining importance because of their ease of operation. The volume concentrations measured can then be used with the instantaneous exhaust flow rate values to obtain mass flow rates of pollutants.

With the recent promulgation of real world …

Kidney Ailment Prediction Under Data Imbalance, Ranaa Mahveen

Graduate Theses, Dissertations, and Problem Reports

Chronic Kidney Disease (CKD) is the leading cause for kidney failure. It is a global health problem affecting approximately 10% of the world population and about 15% of US adults. Chronic Kidney Diseases do not generally show any disease specific symptoms in early stages thus it is hard to detect and prevent such diseases. Early detection and classification are the key factors in managing Chronic Kidney Diseases.

In this thesis, we propose a new machine learning technique for Kidney Ailment Prediction. We focus on two key issues in machine learning, especially in its application to disease prediction. One is related …

Hometracker: A Household Information Feedback System For Food/Energy/Water Metabolism, Nichole Mackey

Dissertations, Master's Theses and Master's Reports

The Food, Energy and Water Conscious (FEWCON) project seeks to understand how food, energy and water (FEW) as independent resources within households are connected. In the main study of the project, intervention messages that link household FEW consumption to equivalent climate consequences are pushed to the households. The goal of the FEWCON study is to determine potential intervention messages that influence household FEW consumption behavior.

A key component of the FEWCON study is a web application named HomeTracker (Household Metabolism Tracker) which collects FEW consumption data within households, then uses this data to select consumption-specific feedback to the homeowners. To …

Building A Classification Model Using Affinity Propagation, Christopher R. Klecker

Electronic Theses and Dissertations

Regular classification of data includes a training set and test set. For example for Naïve Bayes, Artificial Neural Networks, and Support Vector Machines, each classifier employs the whole training set to train itself. This thesis will explore the possibility of using a condensed form of the training set in order to get a comparable classification accuracy. The technique explored in this thesis will use a clustering algorithm to explore with data records can be labeled as exemplar, or a quality of multiple records. For example, is it possible to compress say 50 records into one single record? Can a single …

Eaglebot: A Chatbot Based Multi-Tier Question Answering System For Retrieving Answers From Heterogeneous Sources Using Bert, Muhammad Rana

Electronic Theses and Dissertations

This paper proposes to tackle Question Answering on a specific domain by developing a multi-tier system using three different types of data storage for storing answers. For testing our system on University domain we have used extracted data from Georgia Southern University website. For the task of faster retrieval we have divided our answer data sources into three distinct types and utilized Dialogflow's Natural Language Understanding engine for route selection. We compared different word and sentence embedding techniques for making a semantic question search engine and BERT sentence embedding gave us the best result and for extracting answer from a …

Trapping Aco Applied To Mri Of The Heart, Shannon Lloyd Birchell

UNF Graduate Theses and Dissertations

The research presented here supports the ongoing need for automatic heart volume calculation through the identification of the left and right ventricles in MRI images. The need for automated heart volume calculation stems from the amount of time it takes to manually processes MRI images and required esoteric skill set. There are several methods for region detection such as Deep Neural Networks, Support Vector Machines and Ant Colony Optimization. In this research Ant Colony Optimization (ACO) will be the method of choice due to its efficiency and flexibility. There are many types of ACO algorithms using a variety of heuristics …

On Learning And Visualizing Lexicographic Preference Trees, Ahmed S. Moussa

UNF Graduate Theses and Dissertations

Preferences are very important in research fields such as decision making, recommendersystemsandmarketing. The focus of this thesis is on preferences over combinatorial domains, which are domains of objects configured with categorical attributes. For example, the domain of cars includes car objects that are constructed withvaluesforattributes, such as ‘make’, ‘year’, ‘model’, ‘color’, ‘body type’ and ‘transmission’.Different values can instantiate an attribute. For instance, values for attribute ‘make’canbeHonda, Toyota, Tesla or BMW, and attribute ‘transmission’ can haveautomaticormanual. To this end,thisthesis studiesproblemsonpreference visualization and learning for lexicographic preference trees, graphical preference models that often are compact over complex domains of objects built of …