Digital Commons Network^™

Design Of Experiments For Learning Personalized Visual Preferences Of Occupants In Private Offices, Nimish Awalgaonkar Awalgaonkar, Jie Xiong, Ilias Bilionis, Athanasios Tzempelikos, Panagiota Karava

International High Performance Buildings Conference

Occupants working in offices with controllable shading and lighting systems often perform adaptive actions (raising/lowering shades, increasing/decreasing light levels) in an attempt to restore their preferred state of the room. This preference for different room conditions (states) has been shown to vary from person to person and may be affected by exterior conditions. This paper presents an online data-driven methodology which actively queries a new occupant for learning their personalized visual preferences. Preference is governed by a latent preference relation equivalent to a scalar utility function (the higher the utility, the higher the preference for that state). Information about user …

Methods And Algorithms For Inference Problems In Population Genetics, Jingwen Pei

Doctoral Dissertations

Inference of population history is a central problem of population genetics. The advent of large genetic data brings us not only opportunities on developing more accurate methods for inference problems, but also computational challenges. Thus, we aim at developing accurate method and fast algorithm for problems in population genetics.

Inference of admixture proportions is a classical statistical problem. We particularly focus on the problem of ancestry inference for ancestors. Standard methods implicitly assume that both parents of an individual have the same admixture fraction. However, this is rarely the case in real data. We develop a Hidden Markov Model (HMM) …

Gekko Optimization Suite, Logan Beal, Daniel Hill, Ronald Abraham Martin, John Hedengren

Faculty Publications

This paper introduces GEKKO as an optimization suite for Python. GEKKO specializes in dynamic optimization problems for mixed-integer, nonlinear, and differential algebraic equations (DAE) problems. By blending the approaches of typical algebraic modeling languages (AML) and optimal control packages, GEKKO greatly facilitates the development and application of tools such as nonlinear model predicative control (NMPC), real-time optimization (RTO), moving horizon estimation (MHE), and dynamic simulation. GEKKO is an object-oriented Python library that offers model construction, analysis tools, and visualization of simulation and optimization. In a single package, GEKKO provides model reduction, an object-oriented library for data reconciliation/model predictive control, and …

Identifying Elderlies At Risk Of Becoming More Depressed With Internet-Of-Things, Jiajue Ou, Huiguang Liang, Hwee Xian Tan

Research Collection School Of Computing and Information Systems

Depression in the elderly is common and dangerous. Current methods to monitor elderly depression, however, are costly, time-consuming and inefficient. In this paper, we present a novel depression-monitoring system that infers an elderly’s changes in depression level based on his/her activity patterns, extracted from wireless sensor data. To do so, we build predictive models to learn the relationship between depression level changes and behaviors using historical data. We also deploy the system for a group of elderly, in their homes, and run the experiments for more than one year. Our experimental study gives encouraging results, suggesting that our IoT system …

Non-Destructive Evaluation For Composite Material, Desalegn Temesgen Delelegn

Electrical & Computer Engineering Theses & Dissertations

The Nondestructive Evaluation Sciences Branch (NESB) at the National Aeronautics and Space Administration (NASA) Langley Research Center (LaRC) has conducted impact damage experiments over the past few years with the goal of understanding structural defects in composite materials. The Data Science Team within the NASA LaRC Office of the Chief Information Officer (OCIO) has been working with the Non-Destructive Evaluation (NDE) subject matter experts (SMEs), Dr. Cheryl Rose, from the Structural Mechanics & Concepts Branch and Dr. William Winfree, from the Research Directorate, to develop computer vision solutions using digital image processing and machine learning techniques that can help identify …

The Silencing Power Of Algorithms: How The Facebook News Feed Algorithm Manipulates Users' Perceptions Of Opinion Climates, Callie Jessica Morgan

University Honors Theses

This extended literature review investigates how the architecture and features of the Facebook Newsfeed algorithm, EdgeRank, can inhibit and facilitate the expression of political opinions. This paper will investigate how Elisabeth Noelle-Neumann's theory on public opinion, Spiral of Silence, can be used to assess the Facebook news feed as a political opinion source that actively shapes users' perceptions of minority and majority opinion climates. The feedback loops created by the algorithm's criteria influences users' decisions to self-censor or express their political opinions with interpersonal connections and unfamiliar connections on the site.

Bounding Box Improvement With Reinforcement Learning, Andrew Lewis Cleland

Dissertations and Theses

In this thesis, I explore a reinforcement learning technique for improving bounding box localizations of objects in images. The model takes as input a bounding box already known to overlap an object and aims to improve the fit of the box through a series of transformations that shift the location of the box by translation, or change its size or aspect ratio. Over the course of these actions, the model adapts to new information extracted from the image. This active localization approach contrasts with existing bounding-box regression methods, which extract information from the image only once. I implement, train, and …

An Exploration Of Linear Classifiers For Unsupervised Spiking Neural Networks With Event-Driven Data, Wesley Chavez

Dissertations and Theses

Object recognition in video has seen giant strides in accuracy improvements in the last few years, a testament to the computational capacity of deep convolutional neural networks. However, this computational capacity of software-based neural networks coincides with high power consumption compared to that of some spiking neural networks (SNNs), up to 300,000 times more energy per synaptic event in IBM's TrueNorth chip, for example. SNNs are also well-suited to exploit the precise timing of event-driven image sensors, which transmit asynchronous "events" only when the luminance of a pixel changes above or below a threshold value. The combination of event-based imagers …

Evaluating Beach Water Quality And Dengue Fever Risk Factors By Satellite Remote Sensing And Artificial Neural Networks, Abdiel Elias Laureano-Rosario

USF Tampa Graduate Theses and Dissertations

Climatic variations, together with large-scale environmental forces and human development affect the quality of coastal recreational waters, creating potential risks to human health. These environmental forces, including increased temperature and precipitation, often promote specific vector-borne diseases in the Caribbean and Gulf of Mexico. Human activities affect water quality through discharges from urban areas, including nutrient and other pollutants derived from wastewater systems. Both water quality of recreational beaches and vector-borne diseases can be better managed by understanding their relationship with local environmental forces.

I evaluated how changes in vector-borne diseases and poor recreational water quality were related to specific environmental …

Classification Of One-Year Student Persistence: A Machine Learning Approach, Ben Siebrase

Electronic Theses and Dissertations

Multilayer perceptron neural networks, Gaussian naïve Bayes, and logistic regression classifiers were compared when used to make early predictions regarding one-year college student persistence. Two iterations of each model were built, utilizing a grid search process within 10-fold cross-validation in order to tune model parameters for optimal performance on the classification metrics F-Beta and F-1. The results of logistic regression, the historically favored approach in the domain, were compared to the alternative approaches of multilayer perceptron and naïve Bayes based primarily on FBeta and F-1 score performance on a hold-out dataset. A single logistic regression model was found to perform …

Software Requirements Classification Using Word Embeddings And Convolutional Neural Networks, Vivian Lin Fong

Master's Theses

Software requirements classification, the practice of categorizing requirements by their type or purpose, can improve organization and transparency in the requirements engineering process and thus promote requirement fulfillment and software project completion. Requirements classification automation is a prominent area of research as automation can alleviate the tediousness of manual labeling and loosen its necessity for domain-expertise.

This thesis explores the application of deep learning techniques on software requirements classification, specifically the use of word embeddings for document representation when training a convolutional neural network (CNN). As past research endeavors mainly utilize information retrieval and traditional machine learning techniques, we entertain …

Advanced Malware Detection For Android Platform, Ke Xu

Dissertations and Theses Collection (Open Access)

In the first quarter of 2018, 75.66% of smartphones sales were devices running An- droid. Due to its popularity, cyber-criminals have increasingly targeted this ecosys- tem. Malware running on Android severely violates end users security and privacy, allowing many attacks such as defeating two factor authentication of mobile bank- ing applications, capturing real-time voice calls and leaking sensitive information. In this dissertation, I describe the pieces of work that I have done to effectively de- tect malware on Android platform, i.e., ICC-based malware detection system (IC- CDetector), multi-layer malware detection system (DeepRefiner), and self-evolving and scalable malware detection system (DroidEvolver) …

N-Slope: A One-Class Classification Ensemble For Nuclear Forensics, Justin Kehl

Master's Theses

One-class classification is a specialized form of classification from the field of machine learning. Traditional classification attempts to assign unknowns to known classes, but cannot handle novel unknowns that do not belong to any of the known classes. One-class classification seeks to identify these outliers, while still correctly assigning unknowns to classes appropriately. One-class classification is applied here to the field of nuclear forensics, which is the study and analysis of nuclear material for the purpose of nuclear incident investigations. Nuclear forensics data poses an interesting challenge because false positive identification can prove costly and data is often small, high-dimensional, …

How Copyright Law Can Fix Artificial Intelligence's Implicit Bias Problem, Amanda Levendowski

Georgetown Law Faculty Publications and Other Works

As the use of artificial intelligence (AI) continues to spread, we have seen an increase in examples of AI systems reflecting or exacerbating societal bias, from racist facial recognition to sexist natural language processing. These biases threaten to overshadow AI’s technological gains and potential benefits. While legal and computer science scholars have analyzed many sources of bias, including the unexamined assumptions of its often homogenous creators, flawed algorithms, and incomplete datasets, the role of the law itself has been largely ignored. Yet just as code and culture play significant roles in how AI agents learn about and act in the …

Gaussian Processes With Context-Supported Priors For Active Object Localization, Bruno Jedynak

Portland Institute for Computational Science Publications

We devise an algorithm using a Bayesian optimization framework in conjunction with contextual visual data for the efficient localization of objects in still images. Recent research has demonstrated substantial progress in object localization and related tasks for computer vision. However, many current state-of-the-art object localization procedures still suffer from inaccuracy and inefficiency, in addition to failing to provide a principled and interpretable system amenable to high-level vision tasks. We address these issues with the current research.

Our method encompasses an active search procedure that uses contextual data to generate initial bounding-box proposals for a target object. We train a convolutional …

Cox Processes For Counting By Detection, Purnima Rajan, Yongming Ma, Bruno Jedynak

Portland Institute for Computational Science Publications

In this work, doubly stochastic Poisson (Cox) processes and convolutional neural net (CNN) classifiers are used to estimate the number of instances of an object in an image. Poisson processes are well suited to model events that occur randomly in space, such as the location of objects in an image or the enumeration of objects in a scene. The proposed algorithm selects a subset of bounding boxes in the image domain, then queries them for the presence of the object of interest by running a pre-trained CNN classifier. The resulting observations are then aggregated, and a posterior distribution over the …

Exact Recovery Of Prototypical Atoms Through Dictionary Initialization, Greg Zanotti, Enrico Au-Yeung

DePaul Discoveries

In dictionary learning, a matrix comprised of signals Y is factorized into the product of two matrices: a matrix of prototypical "atoms" D, and a sparse matrix containing coefficients for atoms in D, called X. Dictionary learning finds applications in signal processing, image recognition, and a number of other fields. Many algorithms for solving the dictionary learning problem follow the alternating minimization paradigm; that is, by alternating solving for D and X. In 2014, Agarwal et al. proposed a dictionary initialization procedure that is used before this alternating minimization process. We show that there is a …

Narrowing The Scope Of Failure Prediction Using Targeted Fault Load Injection, Paul L. Jordan, Gilbert L. Peterson, Alan C. Lin, Michael J. Mendenhall, Andrew J. Sellers

Faculty Publications

As society becomes more dependent upon computer systems to perform increasingly critical tasks, ensuring that those systems do not fail becomes increasingly important. Many organizations depend heavily on desktop computers for day-to-day operations. Unfortunately, the software that runs on these computers is written by humans and, as such, is still subject to human error and consequent failure. A natural solution is to use statistical machine learning to predict failure. However, since failure is still a relatively rare event, obtaining labelled training data to train these models is not a trivial task. This work presents new simulated fault-inducing loads that extend …

Theory, Operation, And Application Of Neural Networks, John Andersen

University Honors Theses

This paper examines the history and current state of machine learning. It examines neural networks, theory behind neural networks, how they are implemented, and how they are used. The systems and networks examined have up to three modes of learning. Theory behind machine learning is broken up into three approaches; rule-based, Bayesian, and neural networks. Operation of machine learning algorithms has been enabled by several prevalent libraries in the open source community, as well as various hardware technologies. Due to this surge in resources application developers have been able to apply machine learning in novel ways. An application of machine …

An Investigation Of The Cortical Learning Algorithm, Anthony C. Samaritano

Theses and Dissertations

Pattern recognition and machine learning fields have revolutionized countless industries and applications from biometric security to modern industrial assembly lines. The fields continue to accelerate as faster, more efficient processing hardware becomes commercially available. Despite the accelerated growth of the pattern recognition and machine learning fields, computers still are unable to learn, reason, and perform rudimentary tasks that humans and animals find routine. Animals are able to move fluidly, understand their environment, and maximize their chances of survival through adaptation - animals demonstrate intelligence. A primary argument in this thesis that we have not yet achieved a level of intelligence …

Design Of A Distributed Real-Time E-Health Cyber Ecosystem With Collective Actions: Diagnosis, Dynamic Queueing, And Decision Making, Yanlin Zhou

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

In this thesis, we develop a framework for E-health Cyber Ecosystems, and look into different involved actors. The three interested parties in the ecosystem including patients, doctors, and healthcare providers are discussed in 3 different phases. In Phase 1, machine-learning based modeling and simulation analysis is performed to remotely predict a patient's risk level of having heart diseases in real time. In Phase 2, an online dynamic queueing model is devised to pair doctors with patients having high risk levels (diagnosed in Phase 1) to confirm the risk, and provide help. In Phase 3, a decision making paradigm is proposed …

Machine Learning Applications In Graduation Prediction At The University Of Nevada, Las Vegas, Elliott Collin Ploutz

UNLV Theses, Dissertations, Professional Papers, and Capstones

Graduation rates of four-year institutions are an increasingly important metric to incoming students and for ranking universities. To increase completion rates, universities must analyze available student data to understand trends and factors leading to graduation. Using predictive modeling, incoming students can be assessed as to their likelihood of completing a degree. If students are predicted to be most likely to drop out, interventions can be enacted to increase retention and completion rates.

At the University of Nevada, Las Vegas (UNLV), four-year graduation rates are 15% and six-year graduation rates are 39%. To improve these rates, we have gathered seven years …

Discerning Drivers Of Cancer: Computational Approaches To Somatic Exome Sequencing Data, Runjun Kumar

Arts & Sciences Graduate Student Theses and Dissertations

Paired tumor-normal sequencing of thousands of patient’s exomes has revealed millions of somatic mutations, but functional characterization and clinical decision making are stymied because biologically neutral ‘passenger’ mutations greatly outnumber pathogenic ‘driver’ mutations. Since most mutations will return negative results if tested, conventional resource-intensive experiments are reserved for mutations which are observed in multiple patients or rarer mutations found in well-established cancer genes. Most mutations are therefore never tested, diminishing the potential to discover new mechanisms of cancer development and treatment opportunities. Computational methods that reliably prioritize mutations for testing would greatly increase the translation of sequencing results to clinical …

Malware Image Classification Using Machine Learning With Local Binary Pattern, Jhu-Sin Luo, Dan Lo

Master of Science in Computer Science Theses

Malware classification is a critical part in the cybersecurity.

Traditional methodologies for the malware classification

typically use static analysis and dynamic analysis to identify malware.

In this paper, a malware classification methodology based

on its binary image and extracting local binary pattern (LBP)

features are proposed. First, malware images are reorganized into

3 by 3 grids which is mainly used to extract LBP feature. Second,

the LBP is implemented on the malware images to extract features

in that it is useful in pattern or texture classification. Finally,

Tensorflow, a library for machine learning, is applied to classify

malware images with …

A Multiple Classifier System For Predicting Best-Selling Amazon Products, Michael Kranzlein

Master of Science in Computer Science Theses

In this work, I examine a dataset of Amazon product metadata and propose a heterogeneous multiple classifier system for the task of identifying best-selling products in multiple categories. This system of classifiers consumes the product description and the featured product image as input and feeds them through binary classifiers of the following types: Convolutional Neural Network, Na¨ıve Bayes, Random Forest, Ridge Regression, and Support Vector Machine. While each individual model is largely successful in identifying best-selling products from non best-selling products and from worst-selling products, the multiple classifier system is shown to be stronger than any individual model in the …

A Study Of Altmetrics Using Sentiment Analysis, Luis Arredondo

Honors Capstones

The increasing use of social media has found its way into the realm of professional and scholarly discussion. Whether it’s through negative or positive sentiment (or neither) people are discussing and bringing light to each other’s work in an informal manner through social media. This paper aims to analyze the social media data displayed in altmetrics and display any patterns or correlations between the sentiment data, the platforms, and the impact scores. In our findings we conclude that certain social media platforms hold more opinionated scholarly discussions and also that there is no correlation between social media sentiments and the …

Ai-Human Collaboration Via Eeg, Adam Noack

All College Thesis Program, 2016-2019

As AI becomes ever more competent and integrated into our lives, the issue of AI-human goal misalignment looms larger. This is partially because there is often a rift between what humans explicitly command and what they actually mean. Most contemporary AI systems cannot bridge this gap. In this study we attempted to reconcile the goals of human and machine by using EEG signals from a human to help a simulated agent complete a task.

Face Detection And Recognition Using Moving Window Accumulator With Various Deep Learning Architecture, Anil Kumar Nayak

Computer Science and Engineering Theses - Archive

Recent advancement in the field of Computer Vision and Deep Learning is making object detection and recognition easier. Hence, growing research activities in the field of deep learning are enabling researchers to find new ideas in the area of face detection and recognition. Implementation of such systems has a number of challenges when it comes to the current approaches. In this paper, we have presented a system of Face Detection and Recognition with newly designed deep learning classification models like CNN, Inception and various state of art models like SVM and we also compared the result with FaceNet. Multiple approaches …

From Text Classification To Image Clustering, Problems Less Optimized, Amirhossein Herandi

Computer Science and Engineering Theses - Archive

Machine Learning is thriving. Every industry is using its techniques in some way to improve their efficiency and revenue. However, the focus on research is not divided equally between all of the different areas and problems that this field can tackle and analyze. Currently, Computer Vision is the one area that is being focused very extensively by researchers and companies alike, and as a result has seen an amazing boost in the recent years. This ranges from the well-known problems of classification that use discriminative models all the way to more novel problems that use generative models such as style …

Deep Learning For Recognition Of Objects, Activities, Faces, And Spatio-Temporal Patterns, Amir Ghaderi

Computer Science and Engineering Dissertations - Archive

A popular method in machine learning is Convolutional Neural Network (CNN). CNN had was of high interest to the research community in the 1990s, but after that its popularity receded compared to the Support Vector Machine Support Vector Machine (SVM)[1]. One of the reasons was the relatively lower computational demands of SVM. Training CNNs requires significantly more computational power, time, and data than training SVM. One of the important issues in showing the power of the CNN is the availability of the huge amount of data and introducing big datasets. With increased availability of powerful GPU processing, using several improvements …