Engineering Commons^™

Machine Learning For Stock Prediction Based On Fundamental Analysis, Yuxuan Huang, Luiz Fernando Capretz, Danny Ho

Electrical and Computer Engineering Publications

Application of machine learning for stock prediction is attracting a lot of attention in recent years. A large amount of research has been conducted in this area and multiple existing results have shown that machine learning methods could be successfully used toward stock predicting using stocks’ historical data. Most of these existing approaches have focused on short term prediction using stocks’ historical price and technical indicators. In this paper, we prepared 22 years’ worth of stock quarterly financial data and investigated three machine learning algorithms: Feed-forward Neural Network (FNN), Random Forest (RF) and Adaptive Neural Fuzzy Inference System (ANFIS) for …

Gradient Free Sign Activation Zero One Loss Neural Networks For Adversarially Robust Classification, Yunzhe Xue

Dissertations

The zero-one loss function is less sensitive to outliers than convex surrogate losses such as hinge and cross-entropy. However, as a non-convex function, it has a large number of local minima, andits undifferentiable attribute makes it impossible to use backpropagation, a method widely used in training current state-of-the-art neural networks. When zero-one loss is applied to deep neural networks, the entire training process becomes challenging. On the other hand, a massive non-unique solution probably also brings different decision boundaries when optimizing zero-one loss, making it possible to fight against transferable adversarial examples, which is a common weakness in deep learning …

Towards Adversarial Robustness With 01 Lossmodels, And Novel Convolutional Neural Netsystems For Ultrasound Images, Meiyan Xie

Dissertations

This dissertation investigates adversarial robustness with 01 loss models and a novel convolutional neural net systems for vascular ultrasound images.

In the first part, the dissertation presents stochastic coordinate descent for 01 loss and its sensitivity to adversarial attacks. The study here suggests that 01 loss may be more resilient to adversarial attacks than the hinge loss and further work is required.

In the second part, this dissertation proposes sign activation network with a novel gradient-free stochastic coordinate descent algorithm and its ensembling model. The study here finds that the ensembling model gives a high minimum distortion (as measured by …

Data-Driven Learning For Robot Physical Intelligence, Leidi Zhao

Dissertations

The physical intelligence, which emphasizes physical capabilities such as dexterous manipulation and dynamic mobility, is essential for robots to physically coexist with humans. Much research on robot physical intelligence has achieved success on hyper robot motor capabilities, but mostly through heavily case-specific engineering. Meanwhile, in terms of robot acquiring skills in a ubiquitous manner, robot learning from human demonstration (LfD) has achieved great progress, but still has limitations handling dynamic skills and compound actions. In this dissertation, a composite learning scheme which goes beyond LfD and integrates robot learning from human definition, demonstration, and evaluation is proposed. This method tackles …

Machine Learning For Analog/Mixed-Signal Integrated Circuit Design Automation, Weidong Cao

McKelvey School of Engineering Theses & Dissertations

Analog/mixed-signal (AMS) integrated circuits (ICs) play an essential role in electronic systems by processing analog signals and performing data conversion to bridge the analog physical world and our digital information world.Their ubiquitousness powers diverse applications ranging from smart devices and autonomous cars to crucial infrastructures. Despite such critical importance, conventional design strategies of AMS circuits still follow an expensive and time-consuming manual process and are unable to meet the exponentially-growing productivity demands from industry and satisfy the rapidly-changing design specifications from many emerging applications. Design automation of AMS IC is thus the key to tackling these challenges and has been …

Estimating Homophily In Social Networks Using Dyadic Predictions, George Berry, Antonio Sirianni, Ingmar Weber, Jisun An, Michael Macy

Research Collection School Of Computing and Information Systems

Predictions of node categories are commonly used to estimate homophily and other relational properties in networks. However, little is known about the validity of using predictions for this task. We show that estimating homophily in a network is a problem of predicting categories of dyads (edges) in the graph. Homophily estimates are unbiased when predictions of dyad categories are unbiased. Node-level prediction models, such as the use of names to classify ethnicity or gender, do not generally produce unbiased predictions of dyad categories and therefore produce biased homophily estimates. Bias comes from three sources: sampling bias, correlation between model errors …

Power System Stability Assessment With Supervised Machine Learning, Mirka Mandich

Masters Theses

Power system stability assessment has become an important area of research due to the increased penetration of photovoltaics (PV) in modern power systems. This work explores how supervised machine learning can be used to assess power system stability for the Western Electricity Coordinating Council (WECC) service region as part of the Data-driven Security Assessment for the Multi-Timescale Integrated Dynamics and Scheduling for Solar (MIDAS) project. Data-driven methods offer to improve power flow scheduling through machine learning prediction, enabling better energy resource management and reducing demand on real-time time-domain simulations. Frequency, transient, and small signal stability datasets were created using the …

Graphical Models In Reconstructability Analysis And Bayesian Networks, Marcus Harris, Martin Zwick

Systems Science Faculty Publications and Presentations

Reconstructability Analysis (RA) and Bayesian Networks (BN) are both probabilistic graphical modeling methodologies used in machine learning and artificial intelligence. There are RA models that are statistically equivalent to BN models and there are also models unique to RA and models unique to BN. The primary goal of this paper is to unify these two methodologies via a lattice of structures that offers an expanded set of models to represent complex systems more accurately or more simply. The conceptualization of this lattice also offers a framework for additional innovations beyond what is presented here. Specifically, this paper integrates RA and …

Privacy-Preserving Cloud-Assisted Data Analytics, Wei Bao

Graduate Theses and Dissertations

Nowadays industries are collecting a massive and exponentially growing amount of data that can be utilized to extract useful insights for improving various aspects of our life. Data analytics (e.g., via the use of machine learning) has been extensively applied to make important decisions in various real world applications. However, it is challenging for resource-limited clients to analyze their data in an efficient way when its scale is large. Additionally, the data resources are increasingly distributed among different owners. Nonetheless, users' data may contain private information that needs to be protected.

Cloud computing has become more and more popular in …

Synthetic Aperture Radar Image Recognition Of Armored Vehicles, Christopher Szul [*], Torrey J. Wagner, Brent T. Langhals

Faculty Publications

Synthetic Aperture Radar (SAR) imagery is not affected by weather and allows for day-and-night observations, however it can be difficult to interpret. This work applies classical and neural network machine learning techniques to perform image classification of SAR imagery. The Moving and Stationary Target Acquisition and Recognition dataset from the Air Force Research Laboratory was used, which contained 2,987 total observations of the BMP-2, BTR-70, and T-72 vehicles. Using a 75%/25% train/test split, the classical model achieved an average multi-class image recognition accuracy of 70%, while a convolutional neural network was able to achieve a 97% accuracy with lower model …

Achieving Differential Privacy And Fairness In Machine Learning, Depeng Xu

Graduate Theses and Dissertations

Machine learning algorithms are used to make decisions in various applications, such as recruiting, lending and policing. These algorithms rely on large amounts of sensitive individual information to work properly. Hence, there are sociological concerns about machine learning algorithms on matters like privacy and fairness. Currently, many studies only focus on protecting individual privacy or ensuring fairness of algorithms separately without taking consideration of their connection. However, there are new challenges arising in privacy preserving and fairness-aware machine learning. On one hand, there is fairness within the private model, i.e., how to meet both privacy and fairness requirements simultaneously in …

Mlatticeabc: Generic Lattice Constant Prediction Of Crystal Materials Using Machine Learning, Yuxin Li, Wenhui Yang, Rongzhi Dong, Jianjun Hu

Faculty Publications

Lattice constants such as unit cell edge lengths and plane angles are important parameters of the periodic structures of crystal materials. Predicting crystal lattice constants has wide applications in crystal structure prediction and materials property prediction. Previous work has used machine learning models such as neural networks and support vector machines combined with composition features for lattice constant prediction and has achieved a maximum performance for cubic structures with an average coefficient of determination (R2) of 0.82. Other models tailored for special materials family of a fixed form such as ABX3 perovskites can achieve much higher performance due …

An Inside Vs. Outside Classification System For Wi-Fi Iot Devices, Paul Gralla

Dartmouth College Undergraduate Theses

We are entering an era in which Smart Devices are increasingly integrated into our daily lives. Everyday objects are gaining computational power to interact with their environments and communicate with each other and the world via the Internet. While the integration of such devices offers many potential benefits to their users, it also gives rise to a unique set of challenges. One of those challenges is to detect whether a device belongs to one’s own ecosystem, or to a neighbor – or represents an unexpected adversary. An important part of determining whether a device is friend or adversary is to …

Development Of A Real-Time Single-Lead Single-Beat Frequency-Independent Myocardial Infarction Detector, Harold Martin

FIU Electronic Theses and Dissertations

The central aim of this research is the development and deployment of a novel multilayer machine learning design with unique application for the diagnosis of myocardial infarctions (MIs) from individual heartbeats of single-lead electrocardiograms (EKGs) irrespective of their sampling frequencies over a given range. To the best of our knowledge, this design is the first to attempt inter-patient myocardial infarction detection from individual heartbeats of single-lead (lead II) electrocardiograms that achieves high accuracy and near real-time diagnosis. The processing time of 300 milliseconds to a diagnosis is just at the time range in between extremely fast heartbeats of around 300 …

Indoor Navigation Using Convolutional Neural Networks And Floor Plans, Ricky D. Anderson

Theses and Dissertations

The goal of this thesis is to evaluate a new indoor navigation technique by incorporating floor plans along with monocular camera images into a CNN as a potential means for identifying camera position. Building floor plans are widely available and provide potential information for localizing within the building. This work sets out to determine if a CNN can learn the architectural features of a floor plan and use that information to determine a location. In this work, a simulated indoor data set is created and used to train two CNNs. A classification CNN, which breaks up the floor plan into …

Behavior Modeling For Computer Generated Forces Based On Machine Learning, Zhang Qi, Junjie Zeng, Xu Kai, Qin Long, Quanjun Yin

Journal of System Simulation

Abstract: With the rapid development of Machine Learning, especially deep learning, it has become an important way of modeling Computer Generated Force (CGF) behavior by ML methods, which can overcome the challenges of traditional methods. The existing research and application of three typical learning methods in CGF behavior modeling are discussed, and the effects of introducing learning into different stages of the typical CGF applications are analyzed, and the function and performance requirements of CGF behavior modeling using machine learning are proposed. Four potential research directions in the field for future are proposed.

Holistic Control For Cyber-Physical Systems, Yehan Ma

McKelvey School of Engineering Theses & Dissertations

The Industrial Internet of Things (IIoT) are transforming industries through emerging technologies such as wireless networks, edge computing, and machine learning. However, IIoT technologies are not ready for control systems for industrial automation that demands control performance of physical processes, resiliency to both cyber and physical disturbances, and energy efficiency. To meet the challenges of IIoT-driven control, we propose holistic control as a cyber-physical system (CPS) approach to next-generation industrial automation systems. In contrast to traditional industrial automation systems where computing, communication, and control are managed in isolation, holistic control orchestrates the management of cyber platforms (networks and computing platforms) …

Wind Turbine Parameter Calibration Using Deep Learning Approaches, Rebecca Mccubbin

Electronic Theses and Dissertations

The inertia and damping coefficients are critical to understanding the workings of a wind turbine, especially when it is in a transient state. However, many manufacturers do not provide this information about their turbines, requiring people to estimate these values themselves. This research seeks to design a multilayer perceptron (MLP) that can accurately predict the inertia and damping coefficients using the power data from a turbine during a transient state. To do this, a model of a wind turbine was built in Matlab, and a simulation of a three-phase fault was used to collect realistic fault data to input into …

Continuity Of Chen-Fliess Series For Applications In System Identification And Machine Learning, Rafael Dahmen, W. Steven Gray, Alexander Schmeding

Electrical & Computer Engineering Faculty Publications

Model continuity plays an important role in applications like system identification, adaptive control, and machine learning. This paper provides sufficient conditions under which input-output systems represented by locally convergent Chen-Fliess series are jointly continuous with respect to their generating series and as operators mapping a ball in an Lp-space to a ball in an Lq-space, where p and q are conjugate exponents. The starting point is to introduce a class of topological vector spaces known as Silva spaces to frame the problem and then to employ the concept of a direct limit to describe convergence. The proof of the main …

Improved Online Sequential Extreme Learning Machine: Os-Celm, Olcay Tosun, Recep Eryi̇ği̇t

Turkish Journal of Electrical Engineering and Computer Sciences

Online learning methods (OLM) have been gaining traction as a solution to classification problems because of rapid renewal and fast growth in volume of available data. ELM-based sequential learning (OS-ELM) is one of the most frequently used online learning methodologies partly due to fast training algorithm but suffers from inefficient use of its hidden layers due to the random assignment of the parameters of those layers. In this study, we propose an improved online learning model called online sequential constrained extreme learning machine (OS-CELM), which replaces the random assignment of those parameters with better generalization performance using the CELM method …

Swft: Subbands Wavelet For Local Features Transform Descriptor For Cornealdiseases Diagnosis, Samer Al-Salihi, Sezgi̇n Aydin, Nebras Hussein

Turkish Journal of Electrical Engineering and Computer Sciences

Human cornea is the front see-through shield of the eye. It refracts light onto the retina to induce vision.Therefore, any defect in the cornea may lead to vision disturbance. This deficiency is estimated by sets of topographicalimages measured, and assessed by an ophthalmologist. Consequently, an important priority is the early and accuratediagnosis of diseases that may affect corneal integrity through the use of machine learning algorithms. Images producedby a Pentacam device can be subjected to rotation or some distortion during acquisition; therefore, accurate diagnosisrequires the use of local features in the image. Accordingly, a new algorithm called subbands wavelet for …

Determining And Evaluating New Store Locations Using Remote Sensing Andmachine Learning, Berkan Höke, Zeynep Zerri̇n Turgay, Cem Ünsalan, Hande Küçükaydin

Turkish Journal of Electrical Engineering and Computer Sciences

Decision making for store locations is crucial for retail companies as the profit depends on the location. The key point for correct store location is profit approximation, which is highly dependent on population of the corresponding region, and hence, the volume of the residential area. Thus, estimating building volumes provides insight about the revenue if a new store is about to be opened there. Remote sensing through stereo/tri-stereo satellite images provides wide area coverage as well as adequate resolution for three dimensional reconstruction for volume estimation. We reconstruct 3D map of corresponding region with the help of semiglobal matching and …

An Improved Version Of Multi-View K-Nearest Neighbors (Mvknn) For Multipleview Learning, Eli̇fe Öztürk Kiyak, Derya Bi̇rant, Kökten Ulaş Bi̇rant

Turkish Journal of Electrical Engineering and Computer Sciences

Multi-view learning (MVL) is a special type of machine learning that utilizes more than one views, where views include various descriptions of a given sample. Traditionally, classification algorithms such as k-nearest neighbors (KNN) are designed for learning from single-view data. However, many real-world applications involve datasets with multiple views and each view may contain different and partly independent information, which makes the traditional single-view classification approaches ineffective. Therefore, this article proposes an improved MVL algorithm, called multi-view k-nearest neighbors (MVKNN), based on the existing KNN algorithm. The experimental results conducted in this research show that a significant improvement is achieved …

A New Approach: Semisupervised Ordinal Classification, Ferda Ünal, Derya Bi̇rant, Özlem Şeker

Turkish Journal of Electrical Engineering and Computer Sciences

Semisupervised learning is a type of machine learning technique that constructs a classifier by learning from a small collection of labeled samples and a large collection of unlabeled ones. Although some progress has been made in this research area, the existing semisupervised methods provide a nominal classification task. However, semisupervised learning for ordinal classification is yet to be explored. To bridge the gap, this study combines two concepts ?semisupervised learning? and "ordinal classification" for the categorical class labels for the first time and introduces a new concept of "semisupervised ordinal classification". This paper proposes a new algorithm for semisupervised learning …

A New Classification Method For Encrypted Internet Traffic Using Machine Learning, Mesut Uğurlu, İbrahi̇m Alper Doğru, Recep Si̇nan Arslan

Turkish Journal of Electrical Engineering and Computer Sciences

The rate of internet usage in the world is over 62% and this rate is increasing day by day. With this increase, it becomes important to ensure the confidentiality of the information in the traffic flowing over the internet. Encryption algorithms and protocols are used for this purpose. This situation, which is beneficial for normal users, is also used by attackers to hide. Cyber attackers or hackers gain the ability to bypass security precautions such as IDS/IPS and antivirus systems with using encrypted traffic. Since payload analysis cannot be performed without deciphering the encrypted traffic, existing commercial security solutions fall …

Deep Learning For Turkish Makam Music Composition, İsmai̇l Hakki Parlak, Yalçin Çebi̇, Ci̇han Işikhan, Derya Bi̇rant

Turkish Journal of Electrical Engineering and Computer Sciences

In this paper, we introduce a new deep-learning-based system that can compose structured Turkish makam music (TMM) in the symbolic domain. Presented artificial TMM composer (ATMMC) takes eight initial notes from a human user and completes the rest of the piece. The backbone of the composer system consists of multilayered long short-term memory (LSTM) networks. ATMMC can create pieces in Hicaz and Nihavent makams in Şarkı form, which can be viewed and played with Mus2, a notation software for microtonal music. Statistical analysis shows that pieces composed by ATMMC are approximately 84% similar to training data. ATMMC is an open-source …

Applied Machine Learning In Extrusion-Based Bioprinting, Shuyu Tian

Theses and Dissertations

Optimization of extrusion-based bioprinting (EBB) parameters have been systematically conducted through experimentation. However, the process is time and resource-intensive and not easily translatable across different laboratories. A machine learning (ML) approach to EBB parameter optimization can accelerate this process for laboratories across the field through training using data collected from published literature. In this work, regression-based and classification-based ML models were investigated for their abilities to predict printing outcomes of cell viability and filament diameter for cell-containing alginate and gelatin composite hydrogels. Regression-based models were investigated for their ability to predict suitable extrusion pressure given desired cell viability when keeping …

Texture-Driven Image Clustering In Laser Powder Bed Fusion, Alexander H. Groeger

Browse all Theses and Dissertations

The additive manufacturing (AM) field is striving to identify anomalies in laser powder bed fusion (LPBF) using multi-sensor in-process monitoring paired with machine learning (ML). In-process monitoring can reveal the presence of anomalies but creating a ML classifier requires labeled data. The present work approaches this problem by printing hundreds of Inconel-718 coupons with different processing parameters to capture a wide range of process monitoring imagery with multiple sensor types. Afterwards, the process monitoring images are encoded into feature vectors and clustered to isolate groups in each sensor modality. Four texture representations were learned by training two convolutional neural network …

A Novel Method For Soc Estimation Of Li-Ion Batteries Using A Hybrid Machinelearning Technique, Eymen İpek, Murat Yilmaz

Turkish Journal of Electrical Engineering and Computer Sciences

The battery system is one of the key components of electric vehicles (EV) which has brought groundbreaking technologies. Since modern EVs have mostly Li-ion batteries, they need to be monitored and controlled to achieve safe and high-performance operation. Particularly, the battery management system (BMS) uses complex processing systems that perform measurements, estimation of the battery states, and protection of the system. State of charge (SOC) estimation is a major part of these processes which defines remaining capacity in the battery until the next charging operation as a proportion to the total battery capacity. Since SOC is not a parameter that …

Analysis Of Classifier Weaknesses Based On Patterns And Corrective Methods, Nicholas Skapura

Browse all Theses and Dissertations

Classification is an important branch of machine learning that impacts many areas of modern life. Many classification algorithms (classifiers for short) have been developed. They have highly different levels of sophistication and classification accuracy. Classification problems often have highly different levels of hardness and complexity. Practitioners of classification modeling need better understanding of those algorithms in order to select the optimal algorithm for given classification problems. Researchers of classification need new insight on how given classifiers are weak and how they can be improved by correcting their classification errors. This dissertation introduces new tools and concepts to analyze classifier weakness …