Engineering Commons^™

Evaluating Machine Learning Models For Semantic Segmentation Over Cloud Images For Classification, Harsh Nagarkar

Honors Theses

Due to the increasing number of available approaches nowadays, choosing the most accurate image semantic segmentation model has become hard. The purpose of this research is to find the best-performing image semantic segmentation model for Cloud classification. For the purpose of this study, a data set of cloud images from the Max Planck Institute for meteorology is used. These images were taken from the by two NASA space satellite.Three main models UNet, PSPNet and FPN were used in combination of 4 differ-ent encoder Inception-ResNet-v2, MobileNet-v2, ResNet-34, and ResNet 101. After training all the models in the Mississippi Center for Super …

Knowledge Infused Learning (K-Il): Towards Deep Incorporation Of Knowledge In Deep Learning, Ugur Kursuncu, Manas Gaur, Amit Sheth

Publications

Learning the underlying patterns in data goes beyondinstance-based generalization to external knowledge repre-sented in structured graphs or networks. Deep learning thatprimarily constitutes neural computing stream in AI hasshown significant advances in probabilistically learning la-tent patterns using a multi-layered network of computationalnodes (i.e., neurons/hidden units). Structured knowledge thatunderlies symbolic computing approaches and often supportsreasoning, has also seen significant growth in recent years,in the form of broad-based (e.g., DBPedia, Yago) and do-main, industry or application specific knowledge graphs. Acommon substrate with careful integration of the two willraise opportunities to develop neuro-symbolic learning ap-proaches for AI, where conceptual and probabilistic repre-sentations are combined. …

Algorithm Selection Framework: A Holistic Approach To The Algorithm Selection Problem, Marc W. Chalé

Theses and Dissertations

A holistic approach to the algorithm selection problem is presented. The “algorithm selection framework" uses a combination of user input and meta-data to streamline the algorithm selection for any data analysis task. The framework removes the conjecture of the common trial and error strategy and generates a preference ranked list of recommended analysis techniques. The framework is performed on nine analysis problems. Each of the recommended analysis techniques are implemented on the corresponding data sets. Algorithm performance is assessed using the primary metric of recall and the secondary metric of run time. In six of the problems, the recall of …

Pedestrian Navigation Using Artificial Neural Networks And Classical Filtering Techniques, David J. Ellis

Theses and Dissertations

The objective of this thesis is to explore the improvements achieved through using classical filtering methods with Artificial Neural Network (ANN) for pedestrian navigation techniques. ANN have been improving dramatically in their ability to approximate various functions. These neural network solutions have been able to surpass many classical navigation techniques. However, research using ANN to solve problems appears to be solely focused on the ability of neural networks alone. The combination of ANN with classical filtering methods has the potential to bring beneficial aspects of both techniques to increase accuracy in many different applications. Pedestrian navigation is used as a …

Responsive Economic Model Predictive Control For Next-Generation Manufacturing, Helen Durand

Chemical Engineering and Materials Science Faculty Research Publications

There is an increasing push to make automated systems capable of carrying out tasks which humans perform, such as driving, speech recognition, and anomaly detection. Automated systems, therefore, are increasingly required to respond to unexpected conditions. Two types of unexpected conditions of relevance in the chemical process industries are anomalous conditions and the responses of operators and engineers to controller behavior. Enhancing responsiveness of an advanced control design known as economic model predictive control (EMPC) (which uses predictions of future process behavior to determine an economically optimal manner in which to operate a process) to unexpected conditions of these types …

Machine Learning In Manufacturing: Review, Synthesis, And Theoretical Framework, Ajit Sharma, Zhibo Zhang, Rahul Rai

Business Administration Faculty Research Publications

There has been a paradigmatic shift in manufacturing as computing has transitioned from the programmable to the cognitive computing era. In this paper we present a theoretical framework for understanding this paradigmatic shift in manufacturing and the fast evolving role of artificial intelligence. Policy, Strategic and Operational implications are discussed. Implications for the future of strategy and operations in manufacturing are also discussed. Future research directions are presented.

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 …

Application Of Advanced Algorithms And Statistical Techniques For Weed-Plant Discrimination, Saman Akbar Zadeh

Theses: Doctorates and Masters

Precision agriculture requires automated systems for weed detection as weeds compete with the crop for water, nutrients, and light. The purpose of this study is to investigate the use of machine learning methods to classify weeds/crops in agriculture. Statistical methods, support vector machines, convolutional neural networks (CNNs) are introduced, investigated and optimized as classifiers to provide high accuracy at high vehicular speed for weed detection.

Initially, Support Vector Machine (SVM) algorithms are developed for weed-crop discrimination and their accuracies are compared with a conventional data-aggregation method based on the evaluation of discrete Normalised Difference Vegetation Indices (NDVIs) at two different …

Using Ai And Machine Learning To Indicate Shale Anisotropy And Assist In Completions Design, Cole E. Palmer

Graduate Theses, Dissertations, and Problem Reports (ETD)

Operating companies in the unconventional Marcellus shale play have all faced a similar and problematic issue, while attempting to produce natural gas over the last decade. Companies have quickly realized that not every perforation along their horizontal wells are producing gas. In fact, producing perforations are only ranging from 15%-70% of the total perforations along the horizontal wellbore [1]. This unexplained issue results in millions of dollars in lost revenue per well, in addition to the sunk cost of paying for completions that are not actually yielding any produced gas.

What is causing these perforations to have no produced gas? …

Superconducting Radio-Frequency Cavity Fault Classification Using Machine Learning At Jefferson Laboratory, Chris Tennant, Adam Carpenter, Tom Powers, Anna Shabalina Solopova, Lasitha Vidyaratne, Khan Iftekharuddin

Electrical & Computer Engineering Faculty Publications

We report on the development of machine learning models for classifying C100 superconducting radio-frequency (SRF) cavity faults in the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. CEBAF is a continuous-wave recirculating linac utilizing 418 SRF cavities to accelerate electrons up to 12 GeV through five passes. Of these, 96 cavities (12 cryomodules) are designed with a digital low-level rf system configured such that a cavity fault triggers waveform recordings of 17 rf signals for each of the eight cavities in the cryomodule. Subject matter experts are able to analyze the collected time-series data and identify which of the …

Demonstration Of Visible And Near Infrared Raman Spectrometers And Improved Matched Filter Model For Analysis Of Combined Raman Signals, Alexander Matthew Atkinson

Electrical & Computer Engineering Theses & Dissertations

Raman spectroscopy is a powerful analysis technique that has found applications in fields such as analytical chemistry, planetary sciences, and medical diagnostics. Recent studies have shown that analysis of Raman spectral profiles can be greatly assisted by use of computational models with achievements including high accuracy pure sample classification with imbalanced data sets and detection of ideal sample deviations for pharmaceutical quality control. The adoption of automated methods is a necessary step in streamlining the analysis process as Raman hardware becomes more advanced. Due to limits in the architectures of current machine learning based Raman classification models, transfer from pure …

Recent Trends, Current Research In Cyberpsychology: A Literature Review, Amarjit Kumar Singh, Pawan Kumar Singh

Library Philosophy and Practice (e-journal)

Cyberpsychology refers to the study of the mind and behavior in the context of interactions with technology. It is an emerging branch, which has focused on the psychological aspects connected to the increasing presence and usages of technology in modern lives. This paper traces recent advancement and trends of Cyberpsychology is an emerging domain of knowledge and goes on the give a literature review of the same. An analysis of the recent research and literature covering 300 most relevant research papers from the period of 2012 to 15, August 2019 was conducted to determine and shape the research pattern based …

Mining Semantic Knowledge Graphs To Add Explainability To Black Box Recommender Systems, Mohammed Alshammari, Olfa Nasraoui, Scott Sanders

Faculty and Staff Scholarship

Recommender systems are being increasingly used to predict the preferences of users on online platforms and recommend relevant options that help them cope with information overload. In particular, modern model-based collaborative filtering algorithms, such as latent factor models, are considered state-of-the-art in recommendation systems. Unfortunately, these black box systems lack transparency, as they provide little information about the reasoning behind their predictions. White box systems, in contrast, can, by nature, easily generate explanations. However, their predictions are less accurate than sophisticated black box models. Recent research has demonstrated that explanations are an essential component in bringing the powerful predictions of …

Ambient-Intelligent Decision Support System For Smart Manufacturing, Mahdi Fathi, Marzieh Khakifirooz, Yiannis Ampatzidis, Panos M. Pardalos

BCoE Publications

Ambient Intelligence (AmI) refers to a networked environment of computing devices for implementing a "smart" system. AmI is built using sensors and actuators connected through real-time networks. The data and signals captured from sensors are ambiguous for both human and machine. Artificial Intelligence (AI) is merged into an ambient environment to translate data and signals into a language understandable by human users and can help transform an operational setting from machine-centered to human-centered. However, the implementation of AI technology into an ambient environment requires quantitative modeling approaches to emphasize system requirements analysis and more detailed design specifications. This article tries …

Inverted Cone Convolutional Neural Network For Deboning Mris, Oliver John Palumbo

Theses and Dissertations

Data plenitude is the power but also the bottleneck for data-driven approaches, including neural networks. In particular, Convolutional Neural Networks (CNNs) require an abundant database of training images to achieve a desired high accuracy. Current techniques employed for boosting small datasets are data augmentation and synthetic data generation, which suffer from computational complexity and imprecision compared to original datasets. In this thesis, we intercalate prior knowledge based on the temporal relation between the images in the third dimension. Specifically, we compute the gradient of subsequent images in the dataset to remove extraneous information and highlight subtle variations between the images. …

The Future Robo-Advisor, Catalin Burlacu

MITB Thought Leadership Series

The accelerated digitalisation of both people and business around the world today is having a huge impact on the investment management and advisory space. The addition of new and vastly larger data sets, as well as exponentially more sophisticated analytical tools to turn that data into usable information is constantly changing the way investments are decided on, made and managed.

Receptive Fields Optimization In Deep Learning For Enhanced Interpretability, Diversity, And Resource Efficiency., Babajide Odunitan Ayinde

Electronic Theses and Dissertations

In both supervised and unsupervised learning settings, deep neural networks (DNNs) are known to perform hierarchical and discriminative representation of data. They are capable of automatically extracting excellent hierarchy of features from raw data without the need for manual feature engineering. Over the past few years, the general trend has been that DNNs have grown deeper and larger, amounting to huge number of final parameters and highly nonlinear cascade of features, thus improving the flexibility and accuracy of resulting models. In order to account for the scale, diversity and the difficulty of data DNNs learn from, the architectural complexity and …

Intelligent Machine Learning: Tailor-Making Macromolecules, Yousef Mohammadi, Mohammad Reza Saeb, Alexander Penlidis, Esmaiel Jabbari, Florian J. Stadler, Philippe Zinck, Krzysztof Matyjaszewski

Faculty Publications

Nowadays, polymer reaction engineers seek robust and effective tools to synthesize complex macromolecules with well-defined and desirable microstructural and architectural characteristics. Over the past few decades, several promising approaches, such as controlled living (co)polymerization systems and chain-shuttling reactions have been proposed and widely applied to synthesize rather complex macromolecules with controlled monomer sequences. Despite the unique potential of the newly developed techniques, tailor-making the microstructure of macromolecules by suggesting the most appropriate polymerization recipe still remains a very challenging task. In the current work, two versatile and powerful tools capable of effectively addressing the aforementioned questions have been proposed and …

From Decoder Rings To Deep Fakes: Translating Complex Technologies For Legal Education, Rachel S. Evans, Jason Tubinis

Presentations

“Technological developments are disrupting the practice of law” is a common refrain, but the last few years has seen some particularly complex pieces of technology become the hot new thing in legal tech. This session will look at blockchain, quantum computing, artificial intelligence, and ‘Deep Fakes’ as examples of how librarians can stay abreast of technological developments and inform themselves about their impacts in the legal profession. Then we will look at how to translate the complexities and jargon of these examples into lessons for for-credit courses, one-off informational sessions, or meetings with stakeholders.

Recipe For Disaster, Zac Travis

MFA Thesis Exhibit Catalogs

Today’s rapid advances in algorithmic processes are creating and generating predictions through common applications, including speech recognition, natural language (text) generation, search engine prediction, social media personalization, and product recommendations. These algorithmic processes rapidly sort through streams of computational calculations and personal digital footprints to predict, make decisions, translate, and attempt to mimic human cognitive function as closely as possible. This is known as machine learning.

The project Recipe for Disaster was developed by exploring automation in technology, specifically through the use of machine learning and recurrent neural networks. These algorithmic models feed on large amounts of data as a …

Damage Precursor Identification In Composite Laminates Using Data Driven Approach, Muthu Ram Prabhu Elenchezhian, Vamsee Vadlamudi, Rassel Raihan, Kenneth Reifsnider

Institute of Predictive Performance Methodologies (IPPM-UTARI)-Archive

Composite materials are rapidly being used in commercial aviation and other day to day applications. The individual damage modes in composites are very well understood but it is the interaction of these local damage modes that leads to global failure. In the current research we intend to identify the damage precursors and the initiation of failure events in off axis unidirectional composite laminates loaded in quasi static uniaxial tension by measuring the dielectric response of the material by an in-situ technique using Broadband Dielectric Spectroscopy (BbDS). Using the variation of permittivity with strain, we are able to classify the stages …

Research On Interaction Model Of Hand Tracking Based On Cognitive Theory, Shaobai Zhang, Zhang Teng

Journal of System Simulation

Abstract: This paper aims to solve the problems in a vector integration to endpoint (VITE) model of human reaching and grasping under perturbations of object size, distance and orientation. We discuss how to reduce the numbers of disturbances of three main kinds of components: hand/wrist transport, grip aperture and hand orientation. Based on the achievements of cognitive psychology, and a tracking and cognitive model for operational 3D gestures, this paper proposes a new divide-and-conquer model that is used for indicating current grasping status and to trigger three main kinds of methods of when to start or stop working. The model …

From Situation Cognition Stepped Into Situation Intelligent Cognition, Zhu Feng, Xiaofeng Hu, Wu Lin, Xiaoyuan He, Xuezhi Lü, Liao Ying

Journal of System Simulation

Abstract: Aimed at operational situation cognition and some relevant problems under the background of the Joint-Tactical, some deep researches are carried out in this paper. The concept models of combat situation cognition and situation intelligent cognition are proposed respectively, and some related concepts are clarified. The situation intelligent cognition technology framework is proposed, and five key problems which should be solved are analyzed, and the possible technical routes are given. These research contents and achievements build a foundation for stepping into the situation intelligent cognition from combat situation cognition.

Preliminary Study Of Modeling And Simulation Technology Oriented To Neo-Type Artificial Intelligent Systems, Libo Hu, Xudong Chai, Zhang Lin, Li Tan, Duzheng Qing, Tingyu Lin, Liu Yang

Journal of System Simulation

Abstract: A brief interpretation of the rapidly developing “New Internet+ Big Data+ Artificial Intelligence+” era is given in the paperand the essence and the architectureof neo-type artificial intelligence systems are explained. The meaning of neo-type artificial intelligence system oriented modelling and simulation technology is proposed and the new challenges they are facing are discussed. The research contents and preliminaryresults on neo-type artificial intelligence system oriented modelling and simulation technology are given, which includeneo-type artificial intelligence system oriented modelling/secondary modelling, intelligent simulation computer, smart cloud simulation and intelligent simulation hardware/software supporting system technology, and intelligent simulation system application engineering technology. Several …

Smart Proxy Modeling Of Sacroc Co2-Eor, Gholami Vida, Mohaghegh D. Shahab, Maysami Mohammad

Faculty & Staff Scholarship

Large CO2-enhanced oil recovery (EOR) projects usually contain an abundance of geological and good performance data. While this volume of data leads to robust models, it often results in difficult to manage, slow-running numerical flow models. To dramatically reduce the numerical run-times associated with the traditional simulation techniques, this work investigated the feasibility of using artificial intelligence and machine learning technologies to develop a smart proxy model of the Scurry Area Canyon Reef Operators Committee (SACROC) oilfield, located in the Permian Basin, TX, USA. Smart proxy models can be used to facilitate injection-production optimization for CO2-EOR projects. The use of …

Multi-Sensory Deep Learning Architectures For Slam Dunk Scene Classification, Paul Minogue

Dissertations

Basketball teams at all levels of the game invest a considerable amount of time and effort into collecting, segmenting, and analysing footage from their upcoming opponents previous games. This analysis helps teams identify and exploit the potential weaknesses of their opponents and is commonly cited as one of the key elements required to achieve success in the modern game. The growing importance of this type of analysis has prompted research into the application of computer vision and audio classification techniques to help teams classify scoring sequences and key events using game footage. However, this research tends to focus on classifying …

Distribution Level Building Load Prediction Using Deep Learning, Abdulaziz S. Almalaq

Electronic Theses and Dissertations

Load prediction in distribution grids is an important means to improve energy supply scheduling, reduce the production cost, and support emission reduction. Determining accurate load predictions has become more crucial than ever as electrical load patterns are becoming increasingly complicated due to the versatility of the load profiles, the heterogeneity of individual load consumptions, and the variability of consumer-owned energy resources. However, despite the increase of smart grids technologies and energy conservation research, many challenges remain for accurate load prediction using existing methods. This dissertation investigates how to improve the accuracy of load predictions at the distribution level using artificial …

Should Robots Prosecute And Defend?, Stephen E. Henderson

Faculty Articles

Even when we achieve the ‘holy grail’ of artificial intelligence—machine intelligence that is at least as smart as a human being in every area of thought—there may be classes of decisions for which it is intrinsically important to retain a human in the loop. On the common account of American criminal adjudication, the role of prosecutor seems to include such decisions given the largely unreviewable declination authority, whereas the role of defense counsel would seem fully susceptible of automation. And even for the prosecutor, the benefits of automation might outweigh the intrinsic decision-making loss, given that the ultimate decision—by judge …

Efficient Detection Of Diseases By Feature Engineering Approach From Chest Radiograph, Avishek Mukherjee

Legacy Theses & Dissertations (2009 - 2024)

Deep Learning is the new state-of-the-art technology in Image Processing. We applied Deep Learning techniques for identification of diseases from Radiographs made publicly available by NIH. We applied some Feature Engineering approach to augment the data from Anterior-Posterior position to Posterior-Anterior position and vice-versa for all the diseases, at the same point we suppressed ‘No Finding’ radiographs which contributed to more than 50% (approximately 60,000) of the dataset to top 1000 images. We also prepared a model by adding a huge amount of noise to the augmented data, which if need be can be deployed at rural locations which lack …

Communications Using Deep Learning Techniques, Priti Gopal Pachpande

Legacy Theses & Dissertations (2009 - 2024)

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