Computer Engineering Commons^™

Training Machine Learning Agents In A 3d Game Engine, Diego Calderon

Computer Science and Computer Engineering Undergraduate Honors Theses

Artificial intelligence (AI) and video games benefit from each other. Games provide a challenging domain for testing learning algorithms, and AI provides a framework to designing and implementing intelligent behavior, which reinforces meaningful play. Medium and small studios, and independent game developers, have limited resources to design, implement, and maintain agents with reactive behavior. In this research, we trained agents using machine learning (ML), aiming to find an alternative to expensive traditional algorithms for intelligent behavior used in video games. We use Unity as a game engine to implement the environments and TensorFlow for the neural network training.

Maintainability Analysis Of Mining Trucks With Data Analytics., Abdulgani Kahraman

Electronic Theses and Dissertations

The mining industry is one of the biggest industries in need of a large budget, and current changes in global economic challenges force the industry to reduce its production expenses. One of the biggest expenditures is maintenance. Thanks to the data mining techniques, available historical records of machines’ alarms and signals might be used to predict machine failures. This is crucial because repairing machines after failures is not as efficient as utilizing predictive maintenance. In this case study, the reasons for failures seem to be related to the order of signals or alarms, called events, which come from trucks. The …

Comparative Study Of Deep Learning Models For Network Intrusion Detection, Brian Lee, Sandhya Amaresh, Clifford Green, Daniel Engels

SMU Data Science Review

In this paper, we present a comparative evaluation of deep learning approaches to network intrusion detection. A Network Intrusion Detection System (NIDS) is a critical component of every Internet connected system due to likely attacks from both external and internal sources. A NIDS is used to detect network born attacks such as Denial of Service (DoS) attacks, malware replication, and intruders that are operating within the system. Multiple deep learning approaches have been proposed for intrusion detection systems. We evaluate three models, a vanilla deep neural net (DNN), self-taught learning (STL) approach, and Recurrent Neural Network (RNN) based Long Short …

Distributed Iterative Graph Processing Using Nosql With Data Locality, Ayam Pokhrel

LSU Master's Theses

A tremendous amount of data is generated every day from a wide range of sources such as social networks, sensors, and application logs. Among them, graph data is one type that represents valuable relationships between various entities. Analytics of large graphs has become an essential part of business processes and scientific studies because it leads to deep and meaningful insights into the related domain based on the connections between various entities. However, the optimal processing of large-scale iterative graph computations is very challenging due to the issues like fault tolerance, high memory requirement, parallelization, and scalability. Most of the contemporary …

A Route Less Traveled: Principals’ Perceptions Of Alternative Licensed Cte Teachers, Scott R. Bartholomew, Emma P. Bullock, Louis S. Nadelson

Faculty Publications

The shortage of teachers prepared to teach career and technical education (CTE) courses, or willing to work in certain locations (e.g. inner city, rural), has motivated the desire to explore solutions such as alternative routes to teacher licensure (ARL). Most ARL programs bypass colleges of education and provide a different approach to preparing individuals to teach, leaving many to wonder about the quality and knowledge of these teachers. Specific areas, such as CTE, have seen a large influx of ARL teachers in recent years. We sought to determine school principals’ perceptions of ARL CTE teachers. We found principal support for …

Degradation Modeling And Rul Prediction Using Wiener Process Subject To Multiple Change Points And Unit Heterogeneity, Yuxin Wen, Jianguo Wu, Devashish Das, Tzu-Liang Bill Tseng

Engineering Faculty Articles and Research

Degradation modeling is critical for health condition monitoring and remaining useful life prediction (RUL). The prognostic accuracy highly depends on the capability of modeling the evolution of degradation signals. In many practical applications, however, the degradation signals show multiple phases, where the conventional degradation models are often inadequate. To better characterize the degradation signals of multiple-phase characteristics, we propose a multiple change-point Wiener process as a degradation model. To take into account the between-unit heterogeneity, a fully Bayesian approach is developed where all model parameters are assumed random. At the offline stage, an empirical two-stage process is proposed for model …

Image Processing Applications In Real Life: 2d Fragmented Image And Document Reassembly And Frequency Division Multiplexed Imaging, Houman Kamran Habibkhani

LSU Doctoral Dissertations

In this era of modern technology, image processing is one the most studied disciplines of signal processing and its applications can be found in every aspect of our daily life. In this work three main applications for image processing has been studied.

In chapter 1, frequency division multiplexed imaging (FDMI), a novel idea in the field of computational photography, has been introduced. Using FDMI, multiple images are captured simultaneously in a single shot and can later be extracted from the multiplexed image. This is achieved by spatially modulating the images so that they are placed at different locations in the …

The E-Agriculture Research Landscape In South Africa: A Systematic Literature Review, Alain Nzuzi Kintoki

The African Journal of Information Systems

Despite the growing interest in e-agriculture research in South Africa, academic studies have not sufficiently and deeply investigated the current e-agriculture research trends in the South African context. It is unclear how primary e-agriculture research in South Africa will aid both current and future generations to create new and better ways to transform agricultural development using this modern technology. This study sought to determine the current status of e-agriculture research in the South African context. A systematic literature review was used to gather and analyze data. The results indicate that 17 papers (26.5%) were published during the first two years …

A Sketch-Based Rapid Modeling Method For Crime Scene Presentation, Pu Ren, Wuyang Shui, Jin Liu, Yachun Fan, Wenshuo Zhao, Mingquan Zhou

Journal of Digital Forensics, Security and Law

The reconstruction of crime scene plays an important role in digital forensic application. This article integrates computer graphics, sketch-based retrieval and virtual reality (VR) techniques to develop a low-cost and rapid 3D crime scene presentation approach, which can be used by investigators to analyze and simulate the criminal process. First, we constructed a collection of 3D models for indoor crime scenes using various popular techniques, including laser scanning, image-based modeling and geometric modeling. Second, to quickly obtain an object of interest from the 3D model database, a sketch-based retrieval method was proposed. Finally, a rapid modeling system that integrates our …

Evaluating Classifiers' Optimal Performances Over A Range Of Misclassification Costs By Using Cost-Sensitive Classification, Ramy Al-Saffar

LSU Master's Theses

We believe that using the classification accuracy is not enough to evaluate the performances of classification algorithms. It can be misleading due to overlooking an important element which is the cost if classification is inaccurate. Furthermore, the Receiver Operational Characteristic (ROC) is one of the most popular graphs used to evaluate classifiers performances. However, one of the biggest ROC’s shortcomings is the assumption of equal costs for all misclassified data. Therefore, our goal is to reduce the total cost of decision making by selecting the classifier that has the least total misclassification cost. Nevertheless, the exact misclassification cost is usually …

An Analysis Of Project Setup And Organization In Software, Adam Fitzgibbon

Honors Program: Senior Projects (Public)

The setup of a software project can greatly influence how efficiently software can be built. In this paper, I will be discussing my experiences with various software projects and their setups. These experiences will be used as a basis to draw conclusions on the strengths and weaknesses of specific project setups in certain situations.

Because of the complexity of building software, there isn’t a project setup that works optimally for solving every type of software-based problem. The best approach is to know what variations would best fit the situation and make design decisions from there.

Plant Identification Using Tensorflow, Kody G. Dangtongdee

Computer Engineering

Exploring the diverse nature that is available in San Luis Obispo is one of my favorite activities, however, I often am frustrated that I am unable to identify or learn more about the plants that I appreciate. This project is an attempt at using the concepts of neural networks to create an image classifier which can identify plants. Machine learning is one of the biggest topics in computer science at the moment, and its many uses make it a topic that will continue to be researched for a very long time. Convolutional neural networks are a popular realm of machine …

Skylux Smartphone Controlled Skylight, James A. Green Vi

Computer Engineering

There are numerous electric skylight openers available for purchase for home-use, but the majority of them are remote based, or operated by a wall-unit. Furthermore, these devices are in hard to reach places, so if one were to lose the remote on a remote operated system, the only option is to contact the manufacturer for a new device. As such, my senior project, in collaboration with Colton Sundstrom’s senior project, build upon our existing capstone project in order to allow operation of the Internet of Things (IoT) device over the internet. Our client, Richard Murray, was unsatisfied with the current …

Predicting The Vote Using Legislative Speech, Aditya Budhwar

Master's Theses

As most dedicated observers of voting bodies like the U.S. Supreme Court can attest, it is possible to guess vote outcomes based on statements made during deliberations or questioning by the voting members. In most forms of representative democracy, citizens can actively petition or lobby their representatives, and that often means understanding their intentions to vote for or against an issue of interest. In some U.S. state legislators, professional lobby groups and dedicated press members are highly informed and engaged, but the process is basically closed to ordinary citizens because they do not have enough background and familiarity with the …

Employing A User-Centered Design Process For Cybersecurity Awareness In The Power Grid, Jean C. Scholtz, Lyndsey Franklin, Aditya Ashok, Katya Leblanc, Christopher Bonebrake, Eric Andersen, Michael Cassiadoro

Journal of Human Performance in Extreme Environments

In this paper, we discuss the process we are using in the design and implementation of a tool to improve the situation awareness of cyberattacks in the power grid. We provide details of the steps we have taken to date and describe the steps that still need to be accomplished. The focus of this work is to provide situation awareness of the power grid to staff from different, non-overlapping roles in an electrical transmission organization in order to facilitate an understanding of a possible occurrence of a cyberattack. Our approach follows a user-centered design process and includes determining the types …

Adapt At Semeval-2018 Task 9: Skip-Gram Word Embeddings For Unsupervised Hypernym Discovery In Specialised Corpora, Alfredo Maldonado, Filip Klubicka

Other resources

This paper describes a simple but competitive unsupervised system for hypernym discovery. The system uses skip-gram word embeddings with negative sampling, trained on specialised corpora. Candidate hypernyms for an input word are predicted based on cosine similar- ity scores. Two sets of word embedding mod- els were trained separately on two specialised corpora: a medical corpus and a music indus- try corpus. Our system scored highest in the medical domain among the competing unsu- pervised systems but performed poorly on the music industry domain. Our approach does not depend on any external data other than raw specialised corpora.

Real-Time Assessment And Visual Feedback For Patient Rehabilitation Using Inertial Sensors, Deepa Adinarayanan

ETD Archive

Rehabilitation exercises needs have been continuously increasing and have been projected to increase in future as well based on its demand for aging population, recovering from surgery, injury and illness and the living and working lifestyle of the people. This research aims to tackle one of the most critical issues faced by the exercise administers-Adherence or Non-Adherence to Home Exercise problems especially has been a significant issue resulting in extensive research on the psychological analysis of people involved. In this research, a solution is provided to increase the adherence of such programs through an automated real-time assessment with constant visual …

Continuous Human Activity Tracking Over A Large Area With Multiple Kinect Sensors, Akshat C. Hans

ETD Archive

In recent years, researchers had been inquisitive about the use of technology to enhance the healthcare and wellness of patients with dementia. Dementia symptoms are associated with the decline in thinking skills and memory severe enough to reduce a person’s ability to pay attention and perform daily activities. Progression of dementia can be assessed by monitoring the daily activities of the patients. This thesis encompasses continuous localization and behavioral analysis of patient’s motion pattern over a wide area indoor living space using multiple calibrated Kinect sensors connected over the network. The skeleton data from all the sensor is transferred to …

Using Virtual Reality And Photogrammetry To Enrich 3d Object Identity, Cole Juckette, Heather Richards-Rissetto, Hector Eluid Guerra Aldana, Norman Martinez

Department of Anthropology: Faculty Publications

The creation of digital 3D models for cultural heritage is commonplace. With the advent of efficient and cost effective technologies archaeologists are making a plethora of digital assets. This paper evaluates the identity of 3D digital assets and explores how to enhance or expand that identity by integrating photogrammetric models into VR. We propose that when a digital object acquires spatial context from its virtual surroundings, it gains an identity in relation to that virtual space, the same way that embedding the object with metadata gives it a specific identity through its relationship to other information. We explore this concept …

Blockchain Scalability And Security, Tuyet Duong

Theses and Dissertations

Cryptocurrencies like Bitcoin have proven to be a phenomenal success. The underlying techniques hold huge promise to change the future of financial transactions, and eventually the way people and companies compute, collaborate, and interact. At the same time, the current Bitcoin-like proof-of-work based blockchain systems are facing many challenges. In more detail, a huge amount of energy/electricity is needed for maintaining the Bitcoin blockchain. In addition, their security holds if the majority of the computing power is under the control of honest players. However, this assumption has been seriously challenged recently and Bitcoin-like systems will fail when this assumption is …

Examining A Hate Speech Corpus For Hate Speech Detection And Popularity Prediction, Filip Klubicka, Raquel Fernandez

Other resources

As research on hate speech becomes more and more relevant every day, most of it is still focused on hate speech detection. By attempting to replicate a hate speech detection experiment performed on an existing Twitter corpus annotated for hate speech, we highlight some issues that arise from doing research in the field of hate speech, which is essentially still in its infancy. We take a critical look at the training corpus in order to understand its biases, while also using it to venture beyond hate speech detection and investigate whether it can be used to shed light on other …

Anomaly Inference Based On Heterogeneous Data Sources In An Electrical Distribution System, Yachen Tang

Dissertations, Master's Theses and Master's Reports

Harnessing the heterogeneous data sets would improve system observability. While the current metering infrastructure in distribution network has been utilized for the operational purpose to tackle abnormal events, such as weather-related disturbance, the new normal we face today can be at a greater magnitude. Strengthening the inter-dependencies as well as incorporating new crowd-sourced information can enhance operational aspects such as system reconfigurability under extreme conditions. Such resilience is crucial to the recovery of any catastrophic events. In this dissertation, it is focused on the anomaly of potential foul play within an electrical distribution system, both primary and secondary networks as …

Some Aggregation Operators For Bipolar-Valued Hesitant Fuzzy Information, Florentin Smarandache, Tahir Mahmood, Kifayat Ullah, Qaisar Khan

Branch Mathematics and Statistics Faculty and Staff Publications

In this article we define some aggregation operators for bipolar-valued hesitant fuzzy sets. These operations include bipolar-valued hesitant fuzzy ordered weighted averaging (BPVHFOWA) operator, bipolar-valued hesitant fuzzy ordered weighted geometric (BPVHFOWG) operator and their generalized forms. We also define hybrid aggregation operators and their generalized forms and solved a decision-making problem on these operation.

Quantitative Fine-Grained Human Evaluation Of Machine Translation Systems: A Case Study On English To Croatian, Filip Klubicka, Antonio Toral, Victor Manuel Sanchez-Cartagena

Articles

This paper presents a quantitative fine-grained manual evaluation approach to comparing the performance of different machine translation (MT) systems. We build upon the well-established Multidimensional Quality Metrics (MQM) error taxonomy and implement a novel method that assesses whether the differences in performance for MQM error types between different MT systems are statistically significant. We conduct a case study for English-to- Croatian, a language direction that involves translating into a morphologically rich language, for which we compare three MT systems belonging to different paradigms: pure phrase-based, factored phrase-based and neural. First, we design an MQM-compliant error taxonomy tailored to the relevant …

Is It Worth It? Budget-Related Evaluation Metrics For Model Selection, Filip Klubicka, Giancarlo Salton, John D. Kelleher

Conference papers

Projects that set out to create a linguistic resource often do so by using a machine learning model that pre-annotates or filters the content that goes through to a human annotator, before going into the final version of the resource. However, available budgets are often limited, and the amount of data that is available exceeds the amount of annotation that can be done. Thus, in order to optimize the benefit from the invested human work, we argue that the decision on which predictive model one should employ depends not only on generalized evaluation metrics, such as accuracy and F-score, but …

Application For Position And Load Reference Generation Of A Simulated Mechatronic Chain, Florentin Smarandache, V. Vladareanu, S.B. Cononovici, M. Migdalovici, H. Wang, Y. Feng

Branch Mathematics and Statistics Faculty and Staff Publications

The paper presents the position and load reference generation for a motor stand simulating a mechatronic chain, in this case a three degree of freedom robot leg. The task is accomplished using three PLC controlled motors in position as the robot joint actuators coupled with three controlled in torque, simulating the load at each simulation time-step. The paper briefly discusses the mathematical model and presents the visual interface used in the simulation, which is then to be further integrated into a virtual environment robot control application.

Fundamentals Of Neutrosophic Logic And Sets And Their Role In Artificial Intelligence (Fundamentos De La Lógica Y Los Conjuntos Neutrosóficos Y Su Papel En La Inteligencia Artificial ), Florentin Smarandache, Maykel Leyva-Vazquez

Branch Mathematics and Statistics Faculty and Staff Publications

Neutrosophy is a new branch of philosophy which studies the origin, nature and scope of neutralities. This has formed the basis for a series of mathematical theories that generalize the classical and fuzzy theories such as the neutrosophic sets and the neutrosophic logic. In the paper, the fundamental concepts related to neutrosophy and its antecedents are presented. Additionally, fundamental concepts of artificial intelligence will be defined and how neutrosophy has come to strengthen this discipline.

Modelo De Recomendación Basado En Conocimiento Y Números Svn, Maykel Leyva-Vazquez, Florentin Smarandache

Branch Mathematics and Statistics Faculty and Staff Publications

Recommendation models are useful in the decision-making process that allow the user a set of options that are expected to meet their expectations. Recommendation models are useful in the decision-making process that offer the user a set of options that are expected to meet their SVN expectations to express linguistic terms.

Developing A Recurrent Neural Network With High Accuracy For Binary Sentiment Analysis, Kevin Cunanan

CMC Senior Theses

Sentiment analysis has taken on various machine learning approaches in order to optimize accuracy, precision, and recall. However, Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs) account for the context of a sentence by using previous predictions as additional input for future sentence predictions. Our approach focused on developing an LSTM RNN that could perform binary sentiment analysis for positively and negatively labeled sentences. In collaboration with Mariam Salloum, I developed a collection of programs to classify individual sentences as either positive or negative. This paper additionally looks into machine learning, neural networks, data preprocessing, implementation, and resulting comparisons.

Mobile Cloud Computing: A Comparison Study Of Cuckoo And Aiolos Offloading Frameworks, Inan Kaddour

UNF Graduate Theses and Dissertations

Currently, smart mobile devices are used for more than just calling and texting. They can run complex applications such as GPS, antivirus, and photo editor applications. Smart devices today offer mobility, flexibility, and portability, but they have limited resources and a relatively weak battery. As companies began creating mobile resource intensive and power intensive applications, they have realized that cloud computing was one of the solutions that they could utilize to overcome smart device constraints. Cloud computing helps decrease memory usage and improve battery life. Mobile cloud computing is a current and expanding research area focusing on methods that allow …