Engineering Commons^™

Emotion Forecasting In Dyadic Conversation : Characterizing And Predicting Future Emotion With Audio-Visual Information Using Deep Learning, Sadat Shahriar

Legacy Theses & Dissertations (2009 - 2024)

Emotion forecasting is the task of predicting the future emotion of a speaker, i.e., the emotion label of the future speaking turn–based on the speaker’s past and current audio-visual cues. Emotion forecasting systems require new problem formulations that differ from traditional emotion recognition systems. In this thesis, we first explore two types of forecasting windows(i.e., analysis windows for which the speaker’s emotion is being forecasted): utterance forecasting and time forecasting. Utterance forecasting is based on speaking turns and forecasts what the speaker’s emotion will be after one, two, or three speaking turns. Time forecasting forecasts what the speaker’s emotion will …

Communications Using Deep Learning Techniques, Priti Gopal Pachpande

Legacy Theses & Dissertations (2009 - 2024)

Deep learning (DL) techniques have the potential of making communication systems

Multi-Column Neural Networks And Sparse Coding Novel Techniques In Machine Learning, Ammar O. Hoori

Theses and Dissertations

Accurate and fast machine learning (ML) algorithms are highly vital in artificial intelligence (AI) applications. In complex dataset problems, traditional ML methods such as radial basis function neural network (RBFN), sparse coding (SC) using dictionary learning, and particle swarm optimization (PSO) provide trivial results, large structure, slow training, and/or slow testing. This dissertation introduces four novel ML techniques: the multi-column RBFN network (MCRN), the projected dictionary learning algorithm (PDL) and the multi-column adaptive and non-adaptive particle swarm optimization techniques (MC-APSO and MC-PSO). These novel techniques provide efficient alternatives for traditional ML techniques. Compared to traditional ML techniques, the novel ML …

Novel Applications Of Machine Learning In Bioinformatics, Yi Zhang

Theses and Dissertations--Computer Science

Technological advances in next-generation sequencing and biomedical imaging have led to a rapid increase in biomedical data dimension and acquisition rate, which is challenging the conventional data analysis strategies. Modern machine learning techniques promise to leverage large data sets for finding hidden patterns within them, and for making accurate predictions. This dissertation aims to design novel machine learning-based models to transform biomedical big data into valuable biological insights. The research presented in this dissertation focuses on three bioinformatics domains: splice junction classification, gene regulatory network reconstruction, and lesion detection in mammograms.

A critical step in defining gene structures and mRNA …

Identifying Key Topics Bearing Negative Sentiment On Twitter: Insights Concerning The 2015-2016 Zika Epidemic, Ravali Mamidi, Michele Miller, Tanvi Banerjee, William Romine, Amit Sheth

Publications

Background To understand the public sentiment regarding the Zika virus, social media can be leveraged to understand how positive, negative, and neutral sentiments are expressed in society. Specifically, understanding the characteristics of negative sentiment could help inform federal disease control agencies’ efforts to disseminate relevant information to the public about Zika-related issues.

Objective The purpose of this study was to analyze the public sentiment concerning Zika using posts on Twitter and determine the qualitative characteristics of positive, negative, and neutral sentiments expressed.

Methods Machine learning techniques and algorithms were used to analyze the sentiment of tweets concerning Zika. A supervised …

Less Is More: Beating The Market With Recurrent Reinforcement Learning, Louis Kurt Bernhard Steinmeister

Masters Theses

"Multiple recurrent reinforcement learners were implemented to make trading decisions based on real and freely available macro-economic data. The learning algorithm and different reinforcement functions (the Differential Sharpe Ratio, Differential Downside Deviation Ratio and Returns) were revised and the performances were compared while transaction costs were taken into account. (This is important for practical implementations even though many publications ignore this consideration.) It was assumed that the traders make long-short decisions in the S&P500 with complementary 3-month treasury bill investments. Leveraged positions in the S&P500 were disallowed. Notably, the Differential Sharpe Ratio and the Differential Downside Deviation Ratio are risk …

An Elastic-Net Logistic Regression Approach To Generate Classifiers And Gene Signatures For Types Of Immune Cells And T Helper Cell Subsets, Arezo Torang, Paraag Gupta, David J. Klinke Ii

Faculty & Staff Scholarship

Background: Host immune response is coordinated by a variety of different specialized cell types that vary in time and location. While host immune response can be studied using conventional low-dimensional approaches, advances in transcriptomics analysis may provide a less biased view. Yet, leveraging transcriptomics data to identify immune cell subtypes presents challenges for extracting informative gene signatures hidden within a high dimensional transcriptomics space characterized by low sample numbers with noisy and missing values. To address these challenges, we explore using machine learning methods to select gene subsets and estimate gene coefficients simultaneously. Results: Elastic-net logistic regression, a type of …

Recognition Of Incomplete Objects Based On Synthesis Of Views Using A Geometric Based Local-Global Graphs, Michael Christopher Robbeloth

Browse all Theses and Dissertations

The recognition of single objects is an old research field with many techniques and robust results. The probabilistic recognition of incomplete objects, however, remains an active field with challenging issues associated to shadows, illumination and other visual characteristics. With object incompleteness, we mean missing parts of a known object and not low-resolution images of that object. The employment of various single machine-learning methodologies for accurate classification of the incomplete objects did not provide a robust answer to the challenging problem. In this dissertation, we present a suite of high-level, model-based computer vision techniques encompassing both geometric and machine learning approaches …

Weld Penetration Identification Based On Convolutional Neural Network, Chao Li

Theses and Dissertations--Electrical and Computer Engineering

Weld joint penetration determination is the key factor in welding process control area. Not only has it directly affected the weld joint mechanical properties, like fatigue for example. It also requires much of human intelligence, which either complex modeling or rich of welding experience. Therefore, weld penetration status identification has become the obstacle for intelligent welding system. In this dissertation, an innovative method has been proposed to detect the weld joint penetration status using machine-learning algorithms.

A GTAW welding system is firstly built. Project a dot-structured laser pattern onto the weld pool surface during welding process, the reflected laser pattern …

Applied Deep Learning In Orthopaedics, William Stewart Burton Ii

Electronic Theses and Dissertations

The reemergence of deep learning in recent years has led to its successful application in a wide variety of fields. As a subfield of machine learning, deep learning offers an array of powerful algorithms for data-driven applications. Orthopaedics stands to benefit from the potential of deep learning for advancements in the field. This thesis investigated applications of deep learning for the field of orthopaedics through the development of three distinct projects.

First, algorithms were developed for the automatic segmentation of the structures in the knee from MRI. The resulting algorithms can be used to accurately segment full MRI scans in …

Designing An On-Demand Dynamic Crowdshipping Model And Evaluating Its Ability To Serve Local Retail Delivery In New York City, Shirin Najaf Abadi

Dissertations and Theses

Nowadays city mobility is challenging, mainly in populated metropolitan areas. Growing commute demands, increase in the number of for-hire vehicles, enormous escalation in several intra-city deliveries and limited infrastructure (road capacities), all contribute to mobility challenges. These challenges typically have significant impacts on residents’ quality-of-life particularly from an economic and environmental perspective. Decision-makers have to optimize transportation resources to minimize the system externalities (especially in large-scale metropolitan areas). This thesis focus on the intra-city mobility problems experienced by travelers (in the form of congestion and imbalance taxi resources) and businesses (in the form of last-mile delivery), while taking into consideration …

Reliability Analysis For Systems With Outsourced Components, Zhengwei Hu

Doctoral Dissertations

"The current business model for many industrial firms is to function as system integrators, depending on numerous outsourced components from outside component suppliers. This practice has resulted in tremendous cost savings; it makes system reliability analysis, however, more challenging due to the limited component information available to system designers. The component information is often proprietary to component suppliers. Motivated by the need of system reliability prediction with outsourced components, this work aims to explore feasible ways to accurately predict the system reliability during the system design stage. Four methods are proposed. The first method reconstructs component reliability functions using limited …

Building Shared Knowledge For Eor Technologies: Screening Guideline Constructions, Dashboards, And Advanced Data Analysis, Na Zhang

Doctoral Dissertations

"Successful implementation of enhanced oil recovery (EOR) technology requires comprehensive knowledge and experiences based on existing EOR projects. EOR screening guidelines and EOR reservoir analog are served as such knowledge which are considered as the first step for a reservoir engineer to determine the next step techniques to improve the ultimate oil recovery from their assets. The objective of this research work is to provide better assistance for EOR selection by using fundamental statistics methods and machine learning techniques.

In this dissertation, a total of 977 worldwide EOR projects with the most uniformed, high-quality, and comprehensive information were collected from …

Applications Of Machine Learning In Nuclear Imaging And Radiation Detection, Shaikat Mahmood Galib

Doctoral Dissertations

"The main focus of this work is to use machine learning and data mining techniques to address some challenging problems that arise from nuclear data. Specifically, two problem areas are discussed: nuclear imaging and radiation detection. The techniques to approach these problems are primarily based on a variant of Artificial Neural Network (ANN) called Convolutional Neural Network (CNN), which is one of the most popular forms of 'deep learning' technique.

The first problem is about interpreting and analyzing 3D medical radiation images automatically. A method is developed to identify and quantify deformable image registration (DIR) errors from lung CT scans …

The Efficacy Of Weakly Supervised Learning For Biomedical Imaging, Solomon Lee Hatcher

All ETDs from UAB

This paper aims to investigate the use of unsupervised saliency proposals in extracting suitable prior information from computed tomography scans of the head using image-level labels. The weak supervision learning method obtains useful information about an object’s location and fuses this result with an unsupervised saliency proposal to propagate labels to all related pixels. However, it is shown that unsupervised saliency guidance fails to supply the synthetic annotations with enough coherence to produce suitable segmentation.

Low-Precision Linear Algebra For Neural Networks, Frost Mitchell

Electrical and Computer Engineering

Neural networks have been widely responsible for recent advances in machine learning, powering technologies such as digital assistants and AR photography. LPLANN (Low-Precision Linear Algebra for Neural Networks) is a cross-platform library written in C++ used for implementing neural networks. The software allows users to set specific levels of precision for calculations. Low-precision calculations use advanced parallelization techniques (SIMD, SWAR) to run neural networks at faster rates than full-precision calculations. This library is lightweight enough to run on embedded systems, but also includes optimizations that would provide drastic speedups on a workstation. LPLANN also serves as a testbed for novel …

Computational Modeling Of Trust Factors Using Reinforcement Learning, C. M. Kuzio, A. Dinh, C. Stone, L. Vidyaratne, K. M. Iftekharuddin

Electrical & Computer Engineering Faculty Publications

As machine-learning algorithms continue to expand their scope and approach more ambiguous goals, they may be required to make decisions based on data that is often incomplete, imprecise, and uncertain. The capabilities of these models must, in turn, evolve to meet the increasingly complex challenges associated with the deployment and integration of intelligent systems into modern society. Historical variability in the performance of traditional machine-learning models in dynamic environments leads to ambiguity of trust in decisions made by such algorithms. Consequently, the objective of this work is to develop a novel computational model that effectively quantifies the reliability of autonomous …

Transfer Learning Approach To Multiclass Classification Of Child Facial Expressions, Megan A. Witherow, Manar D. Samad, Khan M. Iftekharuddin

Electrical & Computer Engineering Faculty Publications

The classification of facial expression has been extensively studied using adult facial images which are not appropriate ground truths for classifying facial expressions in children. The state-of-the-art deep learning approaches have been successful in the classification of facial expressions in adults. A deep learning model may be better able to learn the subtle but important features underlying child facial expressions and improve upon the performance of traditional machine learning and feature extraction methods. However, unlike adult data, only a limited number of ground truth images exist for training and validating models for child facial expression classification and there is a …

Exploring And Expanding The One-Pixel Attack, Umairullah Khan, Walt Woods

Maseeh Summer Undergraduate Research Experience

In machine learning research, adversarial examples are normal inputs to a classifier that have been specifically perturbed to cause the model to misclassify the input. These perturbations rarely affect the human readability of an input, even though the model’s output is drastically different. Recent work has demonstrated that image-classifying deep neural networks (DNNs) can be reliably fooled with the modification of a single pixel in the input image, without knowledge of a DNN’s internal parameters. This “one-pixel attack” utilizes an iterative evolutionary optimizer known as differential evolution (DE) to find the most effective pixel to perturb, via the evaluation of …

An Improved Tree Model Based On Ensemble Feature Selection For Classification, Chandralekha M, Shenbagavadivu N

Turkish Journal of Electrical Engineering and Computer Sciences

Researchers train and build specific models to classify the presence and absence of a disease and the accuracy of such classification models is continuously improved. The process of building a model and training depends on the medical data utilized. Various machine learning techniques and tools are used to handle different data with respect to disease types and their clinical conditions. Classification is the most widely used technique to classify disease and the accuracy of the classifier largely depends on the attributes. The choice of the attribute largely affects the diagnosis and performance of the classifier. Due to growing large volumes …

Performance Tuning For Machine Learning-Based Software Development Effort Prediction Models, Egemen Ertuğrul, Zaki̇r Baytar, Çağatay Çatal, Ömer Can Muratli

Turkish Journal of Electrical Engineering and Computer Sciences

Software development effort estimation is a critical activity of the project management process. In this study, machine learning algorithms were investigated in conjunction with feature transformation, feature selection, and parameter tuning techniques to estimate the development effort accurately and a new model was proposed as part of an expert system. We preferred the most general-purpose algorithms, applied parameter optimization technique (GridSearch), feature transformation techniques (binning and one-hot-encoding), and feature selection algorithm (principal component analysis). All the models were trained on the ISBSG datasets and implemented by using the scikit-learn package in the Python language. The proposed model uses a multilayer …

Cloud-Supported Machine Learning System For Context-Aware Adaptive M-Learning, Muhammad Adnan, Asad Habib, Jawad Ashraf, Shafaq Mussadiq

Turkish Journal of Electrical Engineering and Computer Sciences

It is a knotty task to amicably identify the sporadically changing real-world context information of a learner during M-learning processes. Contextual information varies greatly during the learning process. Contextual information that affects the learner during a learning process includes background knowledge, learning time, learning location, and environmental situation. The computer programming skills of learners improve rapidly if they are encouraged to solve real-world programming problems. It is important to guide learners based on their contextual information in order to maximize their learning performance. In this paper, we proposed a cloud-supported machine learning system (CSMLS), which assists learners in learning practical …

Dop: Discover Objects And Paths, A Model For Automated Navigation Andselection In Virtual Environments, Muhammad Raees, Sehat Ullah

Turkish Journal of Electrical Engineering and Computer Sciences

Navigation and selection are the two interaction tasks often needed for the manipulation of an object in a synthetic world. An interface that supports automatic navigation and selection may increase the realism of a virtual reality (VR) system. Such an engrossing interface of a VR system is possible by incorporating machine learning (ML) into the realm of the virtual environment (VE). The use of intelligence in VR systems, however, is a milestone yet to be achieved to make seamless realism in a VE possible. To improve the believability of an intelligent virtual agent (IVA), this research work presents DOP (Discover …

Towards Wearable Blood Pressure Measurement Systems From Biosignals: A Review, Ümi̇t Şentürk, Kemal Polat, İbrahi̇m Yücedağ

Turkish Journal of Electrical Engineering and Computer Sciences

Blood pressure is the pressure by the blood to the vein wall. High blood pressure, which is called silent death, is the cause of nearly 13 % of mortality all over the world. Blood pressure is not only measured in the medical environment, but the blood pressure measurement is also a need for people in their daily life. Blood pressure estimation systems with low error rates have been developed besides the new technologies and algorithms. Blood pressure measurements are differentiated as invasive blood pressure (IBP) measurement and noninvasive blood pressure (NIBP) measurement methods. Although IBP measurement provides the most accurate …

Energy Saving Scheduling In A Fog-Based Iot Application By Bayesian Task Classification Approach, Gholamreza Heydari, Dadmehr Rahbari, Mohsen Nickray

Turkish Journal of Electrical Engineering and Computer Sciences

The Internet of things increases information volume in computer networks and the concept of fog will help us to control this volume more efficiently. Scheduling resources in such an environment would be an NP-Hard problem. This article has studied the concept of scheduling in fog with Bayesian classification which could be applied to gain the task requirements like the processing ones. After classification, virtual machines will be created in accordance with the predicted requirements. The ifogsim simulator has been applied to study our fog-based Bayesian classification scheduling (FBCS) method performance in an EEG tractor application. Algorithms have been evaluated on …

Heart Attack Mortality Prediction: An Application Of Machine Learning Methods, Issam Salman

Turkish Journal of Electrical Engineering and Computer Sciences

The heart is an important organ in the human body, and acute myocardial infarction (AMI) is the leading cause of death in most countries. Researchers are doing a lot of data analysis work to assist doctors in predicting the heart problem. An analysis of the data related to different health problems and its functions can help in predicting the wellness of this organ with a degree of certainty. Our research reported in this paper consists of two main parts. In the first part of the paper, we compare different predictive models of hospital mortality for patients with AMI. All results …

Identifying Criminal Organizations From Their Social Network Structures, Muhammet Serkan Çi̇nar, Burkay Genç, Hayri̇ Sever

Turkish Journal of Electrical Engineering and Computer Sciences

Identification of criminal structures within very large social networks is an essential security feat. By identifying such structures, it may be possible to track, neutralize, and terminate the corresponding criminal organizations before they act. We evaluate the effectiveness of three different methods for classifying an unknown network as terrorist, cocaine, or noncriminal. We consider three methods for the identification of network types: evaluating common social network analysis metrics, modeling with a decision tree, and network motif frequency analysis. The empirical results show that these three methods can provide significant improvements in distinguishing all three network types. We show that these …

Improving Anomaly Detection In Bgp Time-Series Data By New Guide Features And Moderated Feature Selection Algorithm, Mahmoud Hashem, Ahmed Bashandy, Samir Shaheen

Turkish Journal of Electrical Engineering and Computer Sciences

The Internet infrastructure relies on the Border Gateway Protocol (BGP) to provide essential routing information where abnormal routing behavior impairs global Internet connectivity and stability. Hence, employing anomaly detection algorithms is important for improving the performance of BGP routing protocol. In this paper, we propose two algorithms; the first is the guide feature generator (GFG), which generates guide features from traditional features in BGP time-series data using moving regression in combination with smoothed moving average. The second is a modified random forest feature selection algorithm which is employed to automatically select the most dominant features (ASMDF). Our mechanism shows that …

Predicting Cascading Failures In Power Grids Using Machine Learning Algorithms, Rezoan Ahmed Shuvro, Pankaz Das, Majeed M. Hayat, Mitun Talukder

Electrical and Computer Engineering Faculty Research and Publications

Although there has been notable progress in modeling cascading failures in power grids, few works included using machine learning algorithms. In this paper, cascading failures that lead to massive blackouts in power grids are predicted and classified into no, small, and large cascades using machine learning algorithms. Cascading-failure data is generated using a cascading failure simulator framework developed earlier. The data set includes the power grid operating parameters such as loading level, level of load shedding, the capacity of the failed lines, and the topological parameters such as edge betweenness centrality and the average shortest distance for numerous combinations of …

Efficient Machine Learning: Models And Accelerations, Zhe Li

Dissertations - ALL

One of the key enablers of the recent unprecedented success of machine learning is the adoption of very large models. Modern machine learning models typically consist of multiple cascaded layers such as deep neural networks, and at least millions to hundreds of millions of parameters (i.e., weights) for the entire model. The larger-scale model tend to enable the extraction of more complex high-level features, and therefore, lead to a significant improvement of the overall accuracy. On the other side, the layered deep structure and large model sizes also demand to increase computational capability and memory requirements. In order to achieve …