Physical Sciences and Mathematics Commons^™

Carbon Footprint Of Machine Learning Algorithms, Gigi Hsueh

Senior Projects Spring 2020

With the rapid development of machine learning, deep learning has demonstrated superior performance over other types of learning. Research made possible by big data and high-end GPU's enabled those research advances at the expense of computation and environmental costs. This will not only slow down the advancement of deep learning research because not all researchers have access to such expensive hardware, but it also accelerates climate change with increasing carbon emissions. It is essential for machine learning research to obtain high levels of accuracy and efficiency without contributing to global warming. This paper discusses some of current approaches in estimating …

Computer Vision Gesture Recognition For Rock Paper Scissors, Nicholas Hunter

Senior Independent Study Theses

This project implements a human versus computer game of rock-paper-scissors using machine learning and computer vision. Player’s hand gestures are detected using single images with the YOLOv3 object detection system. This provides a generalized detection method which can recognize player moves without the need for a special background or lighting setup. Additionally, past moves are examined in context to predict the most probable next move of the system’s opponent. In this way, the system achieves higher win rates against human opponents than by using a purely random strategy.

Modulation Of Medical Condition Likelihood By Patient History Similarity, Jonathan Turner, Dympna O'Sullivan, Jon Bird

Articles

Introduction: We describe an analysis that modulates the simple population prevalence derived likelihood of a particular condition occurring in an individual by matching the individual with other individuals with similar clinical histories and determining the prevalence of the condition within the matched group.

Methods: We have taken clinical event codes and dates from anonymised longitudinal primary care records for 25,979 patients with 749,053 recorded clinical events. Using a nearest neighbour approach, for each patient, the likelihood of a condition occurring was adjusted from the population prevalence to the prevalence of the condition within those patients with the closest …

A Probabilistic Machine Learning Framework For Cloud Resource Selection On The Cloud, Syeduzzaman Khan

University of the Pacific Theses and Dissertations

The execution of the scientific applications on the Cloud comes with great flexibility, scalability, cost-effectiveness, and substantial computing power. Market-leading Cloud service providers such as Amazon Web service (AWS), Azure, Google Cloud Platform (GCP) offer various general purposes, memory-intensive, and compute-intensive Cloud instances for the execution of scientific applications. The scientific community, especially small research institutions and undergraduate universities, face many hurdles while conducting high-performance computing research in the absence of large dedicated clusters. The Cloud provides a lucrative alternative to dedicated clusters, however a wide range of Cloud computing choices makes the instance selection for the end-users. This thesis …

Model-Based Machine Learning To Identify Clinical Relevance In A High-Resolution Simulation Of Sepsis And Trauma, Zachary H. Silberman Md, Robert Chase Cockrell Phd, Gary An Md

Larner College of Medicine Fourth Year Advanced Integration Teaching/Scholarly Projects

Introduction: Sepsis is a devastating, costly, and complicated disease. It represents the summation of varied host immune responses in a clinical and physiological diagnosis. Despite extensive research, there is no current mediator-directed therapy, nor a biomarker panel able to categorize disease severity or reliably predict outcome. Although still distant from direct clinical translation, dynamic computational and mathematical models of acute systemic inflammation and sepsis are being developed. Although computationally intensive to run and calibrate, agent-based models (ABMs) are one type of model well suited for this. New analytical methods to efficiently extract knowledge from ABMs are needed. Specifically, machine-learning …

Machine Learning Techniques For Quantification Of Knee Segmentation From Mri, Sujeet More, Jimmy Singla, Ahed Abugabah, Ahmad Ali Alzubi

All Works

© 2020 Sujeet More et al. Magnetic resonance imaging (MRI) is precise and efficient for interpreting the soft and hard tissues. Moreover, for the detailed diagnosis of varied diseases such as knee rheumatoid arthritis (RA), segmentation of the knee magnetic resonance image is a challenging and complex task that has been explored broadly. However, the accuracy and reproducibility of segmentation approaches may require prior extraction of tissues from MR images. The advances in computational methods for segmentation are reliant on several parameters such as the complexity of the tissue, quality, and acquisition process involved. This review paper focuses and briefly …

Improving The Accessibility And Transferability Of Machine Learning Algorithms For Identification Of Animals In Camera Trap Images: Mlwic2, Michael A. Tabak, Mohammad S. Norouzzadeh, David W. Wolfson, Erica J. Newton, Raoul K. Boughton, Jacob S. Ivan, Eric Odell, Eric S. Newkirk, Reesa Y. Conrey, Jennifer Stenglein, Fabiola Iannarilli, John Erb, Ryan K. Brook, Amy J. Davis, Jesse Lewis, Daniel P. Walsh, James C. Beasley, Kurt C. Vercauteren, Jeff Clune, Ryan S. Miller

United States Department of Agriculture Wildlife Services: Staff Publications

Motion-activated wildlife cameras (or “camera traps”) are frequently used to remotely and noninvasively observe animals. The vast number of images collected from camera trap projects has prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter-out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. We used 3 million camera …

Implementation Considerations For Mitigating Bias In Supervised Machine Learning, Bardia Bijani Aval

CSB and SJU Distinguished Thesis

Machine Learning (ML) is an important component of computer science and a mainstream way of making sense of large amounts of data. Although the technology is establishing new possibilities in different fields, there are also problems to consider, one of which is bias. Due to the inductive reasoning of ML algorithms in creating mathematical models, the predictions and trends found by the models will never necessarily be true – just more or less probable. Knowing this, it is unreasonable for us to expect the applied deductive reasoning of these models to ever be fully unbiased. Therefore, it is important that …

Predicting Drug Misuse Status Using Machine Learning On Electronic Health Records, Robert Arnold Kania

Master's Theses

Substance misuse is a major problem in the world. in 2014, as many as 52,404 deaths in the US were caused by drug overdoses. in 2001, the monetary cost of drug misuse has been estimated to be 414 billion dollars. in this work, we explore the use of different machine learning algorithms in the prediction of cocaine misuse using structured and unstructured data found in electronic health records. These records contain various attributes that can help with this prediction, including but not limited to chart text data, previous diagnoses of certain diseases and information about the area the patient lives …

Renewable Energy Integration In Distribution System With Artificial Intelligence, Yi Gu

Electronic Theses and Dissertations

With the increasing attention of renewable energy development in distribution power system, artificial intelligence (AI) can play an indispensiable role. In this thesis, a series of artificial intelligence based methods are studied and implemented to further enhance the performance of power system operation and control.

Due to the large volume of heterogeneous data provided by both the customer and the grid side, a big data visualization platform is built to feature out the hidden useful knowledge for smart grid (SG) operation, control and situation awareness. An open source cluster calculation framework with Apache Spark is used to discover big data …

Improving M-Learners' Performance Through Deep Learning Techniques By Leveraging Features Weights, Muhammad Adnan, Asad Habib, Jawad Ashraf, Babar Shah, Gohar Ali

All Works

© 2013 IEEE. Mobile learning (M-learning) has gained tremendous attention in the educational environment in the past decade. For effective M-learning, it is important to create an efficient M-learning model that can identify the exact requirements of mobile learners (M-learners). M-learning model is composed of features that are generated during M-learners' interaction with mobile devices. For an adaptive M-learning model, not only learning features are required, but it is also important to determine how they differ for various M-learners, their weights, and interrelationship. This study proposes a robust and adaptive M-learning model that is based on machine learning and deep …

Detecting Rogue Manipulation Of Smart Home Device Settings, David Zeichick

CCE Theses and Dissertations

Smart home devices control a home’s environmental and security settings. This includes devices that control home thermostats, sprinkler systems, light bulbs, and home appliances. Malicious manipulation of the settings of these devices by an outside adversary has caused emotional distress and could even cause physical harm. For example, researchers have reported that there is a rise in domestic abuse perpetrated via smart home devices; victims have reported their thermostat settings being unwittingly manipulated and being locked out of their house due to their smart lock code being changed. Rapid adoption of smart home devices by consumers has led to an …

Smart Green Communication Protocols Based On Several-Fold Messages Extracted From Common Sequential Patterns In Uavs, Iván García-Magariño, Geraldine Gray, Raquel Lacuesta, Jaime Lloret

Articles

Green communications can be crucial for saving energy in UAVs and enhancing their autonomy. The current work proposes to extract common sequential patterns of communications to gather each common pattern into a single several- fold message with a high-level compression. Since the messages of a pattern are elapsed from each other in time, the current approach performs a machine learning approach for estimating the elapsed times using off-line training. The learned predictive model is applied by each UAV during flight when receiving a several-fold compressed message. We have explored neural networks, linear regression and correlation analyses among others. The current …

Uncovering Host-Microbiome Interactions In Global Systems With Collaborative Programming: A Novel Approach Integrating Social And Data Sciences [Version 1; Peer Review: Awaiting Peer Review], Jenna Oberstaller, Swamy Rakesh Adapa, Guy Dayhoff Ii, Justin Gibbons, Gregory S. Herbert

School of Geosciences Faculty and Staff Publications

Microbiome data are undergoing exponential growth powered by rapid technological advancement. As the scope and depth of microbiome research increases, cross-disciplinary research is urgently needed for interpreting and harnessing the unprecedented data output. However, conventional research settings pose challenges to much-needed interdisciplinary research efforts due to barriers in scientific terminologies, methodology and research-culture. To breach these barriers, our University of South Florida OneHealth Codeathon was designed to be an interactive, hands-on event that solves real-world data problems. The format brought together students, postdocs, faculty, researchers, and clinicians in a uniquely cross-disciplinary, team-focused setting. Teams were formed to encourage equitable distribution …

Using Machine Learning On An Imbalanced Cancer Dataset, James Ekow Arthur

Open Access Theses & Dissertations

With an estimated 1.4 million cancer diagnosis worldwide and the increasing death of cancer patients. It is prudent to investigate methods, approaches and smarter ways of predicting and diagnosing of cancer so that a holistic techniques can be used to curb or reduce false predictions , increase exact predictions and also meticulos prognosis information .

Can a feasible technique be developed for the general problem of prognosis and diagnosis of cancer be developed ?

We will show here that this problem of cancer prognosis and diagnosis can be efficiently tackled with the aid of machine learning techniques and the best, …

Security Techniques For Intelligent Spam Sensing And Anomaly Detection In Online Social Platforms, Monther Aldwairi, Lo'ai Tawalbeh

All Works

Copyright © 2020 Institute of Advanced Engineering and Science. All rights reserved. The recent advances in communication and mobile technologies made it easier to access and share information for most people worldwide. Among the most powerful information spreading platforms are the Online Social Networks (OSN)s that allow Internet-connected users to share different information such as instant messages, tweets, photos, and videos. Adding to that many governmental and private institutions use the OSNs such as Twitter for official announcements. Consequently, there is a tremendous need to provide the required level of security for OSN users. However, there are many challenges due …

Toward Efficient Automation Of Interpretable Machine Learning Boosting, Nathan Neuhaus

All Master's Theses

Developing efficient automated methods for Interpretable Machine Learning (IML) is an important and long-term goal in the field of Artificial Intelligence. Currently the Machine Learning landscape is dominated by Neural Networks (NNs) and Support Vector Machines (SVMs), models which are often highly accurate. Despite high accuracy, such models are essentially “black boxes” and therefore are too risky for situations like healthcare where real lives are at stake. In such situations, so called “glass-box” models, such as Decision Trees (DTs), Bayesian Networks (BNs), and Logic Relational (LR) models are often preferred, however can succumb to accuracy limitations. Unfortunately, having to choose …

Dynamically Updated Diversified Ensemble-Based Approach For Handling Concept Drift, Kanu Goel, Shalini Batra

Turkish Journal of Electrical Engineering and Computer Sciences

Concept drift is the phenomenon where underlying data distribution changes over time unexpectedly. Examining such drifts and getting insight into the executing processes at that instance of time is a big challenge. Prediction models should be capable of handling drifts in scenarios where statistical properties show abrupt changes. Various strategies exist in the literature to deal with such challenging scenarios but the majority of them are limited to the identification of a particular kind of drift pattern. The proposed approach uses online drift detection in a diversified adaptive setting with pruning techniques to formulate a concept drift handling approach, named …

Special Section Guest Editorial: Machine Learning In Optics, Jonathan Howe, Travis Axtell, Khan Iftekharuddin

Electrical & Computer Engineering Faculty Publications

This guest editorial summarizes the Special Section on Machine Learning in Optics.

A Machine Learning System For Glaucoma Detection Using Inexpensive Machine Learning, Jon Kilgannon

West Chester University Master’s Theses

This thesis presents a neural network system which segments images of the retina to calculate the cup-to-disc ratio, one of the diagnostic indicators of the presence or continuing development of glaucoma, a disease of the eye which causes blindness. The neural network is designed to run on commodity hardware and to be run with minimal skill required from the user by packaging the software required to run the network into a Singularity image. The RIGA dataset used to train the network provides images of the retina which have been annotated with the location of the optic cup and disc by …

Probabilistic Machine Learning Using Bayesian Inference, Mayank Pandey

Undergraduate Journal of Mathematical Modeling: One + Two

Machine Learning is a branch of AI (Artificial Intelligence) which expands on the idea of a computational system extending its knowledge about set methodical behaviors from the data that is fed to it to essentially develop analytical skills that can help in identifying patterns and making decisions with little to no participation of a real human being. Computer algorithms help in gaining experience to improve the facility over time for use by both consumers and corporations. In today’s technologically advanced world, Machine Learning has given us self-driving cars, speech recognition software, and AI agents like Siri and Google assistant. This …

Evaluating An Ordinal Output Using Data Modeling, Algorithmic Modeling, And Numerical Analysis, Martin Keagan Wynne Brown

Murray State Theses and Dissertations

Data and algorithmic modeling are two different approaches used in predictive analytics. The models discussed from these two approaches include the proportional odds logit model (POLR), the vector generalized linear model (VGLM), the classification and regression tree model (CART), and the random forests model (RF). Patterns in the data were analyzed using trigonometric polynomial approximations and Fast Fourier Transforms. Predictive modeling is used frequently in statistics and data science to find the relationship between the explanatory (input) variables and a response (output) variable. Both approaches prove advantageous in different cases depending on the data set. In our case, the data …

Experiments On The Neural Network Approach To The Handwritten Digit Classification Problem, William Meissner

Electronic Theses and Dissertations

When the MNIST dataset was introduced in 1998, training a network was a multiple week problem in order to receive results far less accurate than an average CPU can produce within a couple of hours today. While this indicates that training a network on such a dataset is not the complicated problem it may have been twenty years ago, the MNIST dataset makes a good tool for study and testing with beginner and medium complexity neural networks. This paper follows along with the work presented in the online textbook “Neural Networks and Deep Learning” by Michael Nielson and an updated …

Comparing Predictive Performance Of Statistical Learning Models On Medical Data, Francis Biney

Open Access Theses & Dissertations

This work investigates the predictive performance of 10 Machine learning models on three medical data including Breast cancer, Heart disease and Prostate cancer. Furthermore, we use the models to identify risk factors that contribute significantly to these diseases.

The models considered include; Logistic regression with L1 and L_2 penalties, Principal component logistic regression(PCR-LR), Partial least squares logistic regression(PLS-LR), Multivariate adaptive regression splines(MARS), Support vector machine with Radial Basis Kernel (SVM-RBK), Random Forest(RF), Gradient Boosting Machines(GBM), Elastic Net (Enet) and Feedforward Neural Network(FFNN). The models were grouped according to their similarities and learning style; i) Linear regularized models: LR-Lasso, LR-Ridge and …

Automated Recognition Of Facial Affect Using Deep Neural Networks, Behzad Hasani

Electronic Theses and Dissertations

Automated Facial Expression Recognition (FER) has been a topic of study in the field of computer vision and machine learning for decades. In spite of efforts made to improve the accuracy of FER systems, existing methods still are not generalizable and accurate enough for use in real-world applications. Many of the traditional methods use hand-crafted (a.k.a. engineered) features for representation of facial images. However, these methods often require rigorous hyper-parameter tuning to achieve favorable results.

Recently, Deep Neural Networks (DNNs) have shown to outperform traditional methods in visual object recognition. DNNs require huge data as well as powerful computing units …

A Hierarchical Temporal Memory Sequence Classifier For Streaming Data, Jeffrey Barnett

CCE Theses and Dissertations

Real-world data streams often contain concept drift and noise. Additionally, it is often the case that due to their very nature, these real-world data streams also include temporal dependencies between data. Classifying data streams with one or more of these characteristics is exceptionally challenging. Classification of data within data streams is currently the primary focus of research efforts in many fields (i.e., intrusion detection, data mining, machine learning). Hierarchical Temporal Memory (HTM) is a type of sequence memory that exhibits some of the predictive and anomaly detection properties of the neocortex. HTM algorithms conduct training through exposure to a stream …

Development Of Machine Learning Models To Predict The Online Impact Of Research, Mohammed Murtuza Shahzad Syed

Graduate Research Theses & Dissertations

Scientific research is being increasingly shared online in a way such that there is a need to develop methodologies to measure the impact of specific papers in ways that go beyond traditional indicators of scholarly citations and beyond the scholarly community. In this thesis, new machine learning models are developed to measure and predict the impact ofresearch in the online context. The extent to which research papers are mentioned on social media platforms, i.e., their online sustainability, indicates the public's interest in and perhaps even the level of understanding of scientific topics. A research paper having a long lifespan, i.e., …

Mlcocoa: A Machine Learning-Based Congestion Control For Coap, Alper Kami̇l Demi̇r, Fati̇h Abut

Turkish Journal of Electrical Engineering and Computer Sciences

Internet of Things (IoT) is a technological invention that has the potential to impact on how we live and how we work by connecting any device to the Internet. Consequently, a vast amount of novel applications will enhance our lives. Internet Engineering Task Force (IETF) standardized the Constrained Application Protocol (CoAP) to accommodate the application layer and network congestion needs of such IoT networks. CoAP is designed to be very simple where it employs a genuine congestion control (CC) mechanism, named as default CoAP CC leveraging basic binary exponential backoff. Yet efficient, default CoAP CC does not always utilize the …

Pretraining Deep Learning Models For Natural Language Understanding, Han Shao

Honors Papers

Since the first bidirectional deep learn- ing model for natural language understanding, BERT, emerged in 2018, researchers have started to study and use pretrained bidirectional autoencoding or autoregressive models to solve language problems. In this project, I conducted research to fully understand BERT and XLNet and applied their pretrained models to two language tasks: reading comprehension (RACE) and part-of-speech tagging (The Penn Treebank). After experimenting with those released models, I implemented my own version of ELECTRA, a pretrained text encoder as a discriminator instead of a generator to improve compute-efficiency, with BERT as its underlying architecture. To reduce the number …

A Deep Learning Approach To Mapping Irrigation: U-Net Irrmapper, Thomas Henry Colligan Iv

Graduate Student Theses, Dissertations, & Professional Papers

Accurate maps of irrigation are essential for understanding and managing water resources in light of a warming climate. We present a new method for mapping irrigation and apply it to the state of Montana over the years 2000-2019. The method is based on an ensemble of convolutional neural networks that only rely on raw Landsat surface reflectance data. The ensemble of networks method learns to mask clouds and ignore Landsat 7 scan-line failures without supervision, reducing the need for preprocessing data or feature engineering. Unlike other approaches to mapping irrigation, the method doesn't use other mapping products like the Cropland …