Engineering Commons^™

Classification Of Images Based On Pixels That Represent A Small Part Of The Scene. A Case Applied To Microaneurysms In Fundus Retina Images, Pablo F. Ordonez, Pablo F. Ordonez

Master of Science in Computer Science Theses

Convolutional Neural Networks (CNNs), the state of the art in image classification, have proven to be as effective as an ophthalmologist, when detecting Referable Diabetic Retinopathy (RDR). Having a size of less than 1\% of the total image, microaneurysms are early lesions in DR that are difficult to classify. The purpose of this thesis is to improve the accuracy of detection of microaneurysms using a model that includes two CNNs with different input image sizes, 60x60 and 420x420 pixels. These models were trained using the Kaggle and Messidor datasets and tested independently against the Kaggle dataset, showing a sensitivity of …

Feature Selection And Improving Classification Performance For Malware Detection, Carlos A. Cepeda Mora

Master of Science in Computer Science Theses

The ubiquitous advance of technology has been conducive to the proliferation of cyber threats, resulting in attacks that have grown exponentially. Consequently, researchers have developed models based on machine learning algorithms for detecting malware. However, these methods require significant amount of extracted features for correct malware classification, making that feature extraction, training, and testing take significant time; even more, it has been unexplored which are the most important features for accomplish the correct classification.

In this Thesis, it is created and analyzed a dataset of malware and clean files (goodware) from the static and dynamic features provided by the online …

Deep Learning Approach For Intrusion Detection System (Ids) In The Internet Of Things (Iot) Network Using Gated Recurrent Neural Networks (Gru), Manoj Kumar Putchala

Browse all Theses and Dissertations

The Internet of Things (IoT) is a complex paradigm where billions of devices are connected to a network. These connected devices form an intelligent system of systems that share the data without human-to-computer or human-to-human interaction. These systems extract meaningful data that can transform human lives, businesses, and the world in significant ways. However, the reality of IoT is prone to countless cyber-attacks in the extremely hostile environment like the internet. The recent hack of 2014 Jeep Cherokee, iStan pacemaker, and a German steel plant are a few notable security breaches. To secure an IoT system, the traditional high-end security …

Internal Medicine, Keroles Hakem, Robert Trachy, Khanh Tran

Capstone Design Expo Posters

Our objective was to develop a model to predict the length of stay of patients using data from MCV. We conducted our analysis using a dataset of over 130,000 patients described by 66 features. The features contained clinical characteristics (e.g. diagnosis), facility characteristics (e.g. bed type), and socioeconomic characteristics (e.g. insurance type). Our study was focused on patients that stayed in the hospital. To cope with data imperfections, such as missing data, we applied data cleaning methods. Using learned domain knowledge, we identified 9 features to build our predictive models: admit source, primary insurance, discharge disposition, admit unit, iso result, …

Using Natural Language Processing And Machine Learning Techniques To Characterize Configuration Bug Reports: A Study, Wei Wen

Theses and Dissertations--Computer Science

In this study, a tool is developed that achieves two purposes: (1) given bug reports, it identifies configuration bug reports from non-configuration bug reports; (2) once a bug report is identified to be a configuration bug report, the tool finds out what specific configuration option the bug report is associated.

This study starts with a review of related works that used machine learning tools to solve software bug and bug report related issues. It then discusses the natural language processing and machine learning techniques. Afterwards, the development process of the proposed tool is described in detail, including the motivation, the …

Multi-Class Classification Of Textual Data: Detection And Mitigation Of Cheating In Massively Multiplayer Online Role Playing Games, Naga Sai Nikhil Maguluri

Browse all Theses and Dissertations

The success of any multiplayer game depends on the player’s experience. Cheating/Hacking undermines the player’s experience and thus the success of that game. Cheaters, who use hacks, bots or trainers are ruining the gaming experience of a player and are making him leave the game. As the video game industry is a constantly increasing multibillion dollar economy, it is crucial to assure and maintain a state of security. Players reflect their gaming experience in one of the following places: multiplayer chat, game reviews, and social media. This thesis is an exploratory study where our goal is to experiment and propose …

Optimized Multilayer Perceptron With Dynamic Learning Rate To Classify Breast Microwave Tomography Image, Chulwoo Pack

Electronic Theses and Dissertations

Most recently developed Computer Aided Diagnosis (CAD) systems and their related research is based on medical images that are usually obtained through conventional imaging techniques such as Magnetic Resonance Imaging (MRI), x-ray mammography, and ultrasound. With the development of a new imaging technology called Microwave Tomography Imaging (MTI), it has become inevitable to develop a CAD system that can show promising performance using new format of data. The platform can have a flexibility on its input by adopting Artificial Neural Network (ANN) as a classifier. Among the various phases of CAD system, we have focused on optimizing the classification phase …

Investigating The Impact Of Unsupervised Feature-Extraction From Multi-Wavelength Image Data For Photometric Classification Of Stars, Galaxies And Qsos, Annika Lindh

Conference papers

Accurate classification of astronomical objects currently relies on spectroscopic data. Acquiring this data is time-consuming and expensive compared to photometric data. Hence, improving the accuracy of photometric classification could lead to far better coverage and faster classification pipelines. This paper investigates the benefit of using unsupervised feature-extraction from multi-wavelength image data for photometric classification of stars, galaxies and QSOs. An unsupervised Deep Belief Network is used, giving the model a higher level of interpretability thanks to its generative nature and layer-wise training. A Random Forest classifier is used to measure the contribution of the novel features compared to a set …

Activist: A New Framework For Dataset Labelling, Jack O'Neill, Sarah Jane Delany, Brian Mac Namee

Conference papers

Acquiring labels for large datasets can be a costly and time-consuming process. This has motivated the development of the semi-supervised learning problem domain, which makes use of unlabelled data — in conjunction with a small amount of labelled data — to infer the correct labels of a partially labelled dataset. Active Learning is one of the most successful approaches to semi-supervised learning, and has been shown to reduce the cost and time taken to produce a fully labelled dataset. In this paper we present Activist; a free, online, state-of-the-art platform which leverages active learning techniques to improve the efficiency of …

Significant Permission Identification For Android Malware Detection, Lichao Sun

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

A recent report indicates that a newly developed malicious app for Android is introduced every 11 seconds. To combat this alarming rate of malware creation, we need a scalable malware detection approach that is effective and efficient. In this thesis, we introduce SigPID, a malware detection system based on permission analysis to cope with the rapid increase in the number of Android malware. Instead of analyzing all 135 Android permissions, our approach applies 3-level pruning by mining the permission data to identify only significant permissions that can be effective in distinguishing benign and malicious apps. Based on the identified significant …

Scale Up Bayesian Network Learning, Xiannian Fan

Dissertations, Theses, and Capstone Projects

Bayesian networks are widely used graphical models which represent uncertain relations between the random variables in a domain compactly and intuitively. The first step of applying Bayesian networks to real-word problems is typically building the network structure. Optimal structure learning via score-and-search has become an active research topic in recent years. In this context, a scoring function is used to measure the goodness of fit of a structure to given data, and the goal is to find the structure which optimizes the scoring function. The problem has been viewed as a shortest path problem, and has been shown to be …

Towards Building An Intelligent Integrated Multi-Mode Time Diary Survey Framework, Hariharan Arunachalam

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

Enabling true responses is an important characteristic in surveys; where the responses are free from bias and satisficing. In this thesis, we examine the current state of surveys, briefly touching upon questionnaire surveys, and then on time diary surveys (TDS). TDS are open-ended conversational surveys of a free-form nature with both, the interviewer and the respondent, playing a part in its progress and successful completion. With limited research available on how intelligent and assistive components can affect TDS respondents, we explore ways in which intelligent systems such as Computer Adaptive Testing, Intelligent Tutoring Systems, Recommender Systems, and Decision Support Systems …

A Closed Loop Research Platform That Enables Dynamic Control Of Wing Gait Patterns In A Vertically Constrained Flapping Wing - Micro Air Vehicle, Hermanus Van Botha

Browse all Theses and Dissertations

Research in Flapping Wing - Micro Air Vehicles(FW-MAVs) has been growing in recent years. Work ranging from mechanical designs to adaptive control algorithms are being developed in pursuit of mimicking natural flight. FW-MAV technology can be applied in a variety of use cases such a military application and surveillance, studying natural ecological systems, and hobbyist commercialization. Recent work has produced small scale FW-MAVs that are capable of hovering and maneuvering. Researchers control maneuvering in various ways, some of which involve making small adjustments to the core wing motion patterns (wing gaits) which determine how the wings flap. Adaptive control algorithms …

Improving Understandability And Uncertainty Modeling Of Data Using Fuzzy Logic Systems, Dumidu S. Wijayasekara

Theses and Dissertations

The need for automation, optimality and efficiency has made modern day control and monitoring systems extremely complex and data abundant. However, the complexity of the systems and the abundance of raw data has reduced the understandability and interpretability of data which results in a reduced state awareness of the system. Furthermore, different levels of uncertainty introduced by sensors and actuators make interpreting and accurately manipulating systems difficult. Classical mathematical methods lack the capability to capture human knowledge and increase understandability while modeling such uncertainty.

Fuzzy Logic has been shown to alleviate both these problems by introducing logic based on vague …

Neuron Clustering For Mitigating Catastrophic Forgetting In Supervised And Reinforcement Learning, Benjamin Frederick Goodrich

Doctoral Dissertations

Neural networks have had many great successes in recent years, particularly with the advent of deep learning and many novel training techniques. One issue that has affected neural networks and prevented them from performing well in more realistic online environments is that of catastrophic forgetting. Catastrophic forgetting affects supervised learning systems when input samples are temporally correlated or are non-stationary. However, most real-world problems are non-stationary in nature, resulting in prolonged periods of time separating inputs drawn from different regions of the input space.

Reinforcement learning represents a worst-case scenario when it comes to precipitating catastrophic forgetting in neural networks. …

Performance Analysis Of Hybrid Algorithms For Lossless Compression Of Climate Data, Bharath Chandra Mummadisetty

UNLV Theses, Dissertations, Professional Papers, and Capstones

Climate data is very important and at the same time, voluminous. Every minute a new entry is recorded for different climate parameters in climate databases around the world. Given the explosive growth of data that needs to be transmitted and stored, there is a necessity to focus on developing better transmission and storage technologies. Data compression is known to be a viable and effective solution to reduce bandwidth and storage requirements of bulk data. So, the goal is to develop the best compression methods for climate data.

The methodology used is based on predictive analysis. The focus is to implement …

Computational Analysis Of Neutron Scattering Data, Benjamin Walter Martin

Doctoral Dissertations

This work explores potential methods for use in the detection and classification of defects within crystal structures via analysis of diffuse scattering data generated by single crystal neutron scattering experiments. The proposed defect detection methodology uses machine learning and image processing techniques to perform image texture analysis on neutron diffraction patterns generated by neutron scattering simulations. Once the methodology is presented, it is tested via a series of defect detection problems of increasing difficulty which utilize neutron scattering data simulated by a number of simulation techniques. As the problem difficulty is increased, the defect detection methodology is refined in order …

Contrast Pattern Aided Regression And Classification, Vahid Taslimitehrani

Browse all Theses and Dissertations

Regression and classification techniques play an essential role in many data mining tasks and have broad applications. However, most of the state-of-the-art regression and classification techniques are often unable to adequately model the interactions among predictor variables in highly heterogeneous datasets. New techniques that can effectively model such complex and heterogeneous structures are needed to significantly improve prediction accuracy. In this dissertation, we propose a novel type of accurate and interpretable regression and classification models, named as Pattern Aided Regression (PXR) and Pattern Aided Classification (PXC) respectively. Both PXR and PXC rely on identifying regions in the data space where …

Optimizing Parallel Belief Propagation In Junction Trees Using Regression, Lu Zheng, Ole J. Mengshoel

Ole J Mengshoel

Mobile Computing: Challenges And Opportunities For Autonomy And Feedback, Ole J. Mengshoel, Bob Iannucci, Abe Ishihara

Ole J Mengshoel

Learning With An Insufficient Supply Of Data Via Knowledge Transfer And Sharing, Samir Al-Stouhi

Wayne State University Dissertations

As machine learning methods extend to more complex and diverse set of problems, situations arise where the complexity and availability of data presents a situation where the information source is not "adequate" to generate a representative hypothesis. Learning from multiple sources of data is a promising research direction as researchers leverage ever more diverse sources of information. Since data is not readily available, knowledge has to be transferred from other sources and new methods (both supervised and un-supervised) have to be developed to selectively share and transfer knowledge. In this dissertation, we present both supervised and un-supervised techniques to tackle …

Deep Machine Learning With Spatio-Temporal Inference, Thomas Paul Karnowski

Doctoral Dissertations

Deep Machine Learning (DML) refers to methods which utilize hierarchies of more than one or two layers of computational elements to achieve learning. DML may draw upon biomemetic models, or may be simply biologically-inspired. Regardless, these architectures seek to employ hierarchical processing as means of mimicking the ability of the human brain to process a myriad of sensory data and make meaningful decisions based on this data. In this dissertation we present a novel DML architecture which is biologically-inspired in that (1) all processing is performed hierarchically; (2) all processing units are identical; and (3) processing captures both spatial and …

Falconet: Force-Feedback Approach For Learning From Coaching And Observation Using Natural And Experiential Training, Gary Stein

Electronic Theses and Dissertations

Building an intelligent agent model from scratch is a difficult task. Thus, it would be preferable to have an automated process perform this task. There have been many manual and automatic techniques, however, each of these has various issues with obtaining, organizing, or making use of the data. Additionally, it can be difficult to get perfect data or, once the data is obtained, impractical to get a human subject to explain why some action was performed. Because of these problems, machine learning from observation emerged to produce agent models based on observational data. Learning from observation uses unobtrusive and purely …

Contextualizing Observational Data For Modeling Human Performance, Viet Trinh

Electronic Theses and Dissertations

This research focuses on the ability to contextualize observed human behaviors in efforts to automate the process of tactical human performance modeling through learning from observations. This effort to contextualize human behavior is aimed at minimizing the role and involvement of the knowledge engineers required in building intelligent Context-based Reasoning (CxBR) agents. More specifically, the goal is to automatically discover the context in which a human actor is situated when performing a mission to facilitate the learning of such CxBR models. This research is derived from the contextualization problem left behind in Fernlund's research on using the Genetic Context Learner …

A Reinforcement Learning Technique For Enhancing Human Behavior Models In A Context-Based Architecture, David Aihe

Electronic Theses and Dissertations

A reinforcement-learning technique for enhancing human behavior models in a context-based learning architecture is presented. Prior to the introduction of this technique, human models built and developed in a Context-Based reasoning framework lacked learning capabilities. As such, their performance and quality of behavior was always limited by what the subject matter expert whose knowledge is modeled was able to articulate or demonstrate. Results from experiments performed show that subject matter experts are prone to making errors and at times they lack information on situations that are inherently necessary for the human models to behave appropriately and optimally in those situations. …

An Adaptive Multiobjective Evolutionary Approach To Optimize Artmap Neural Networks, Assem Kaylani

Electronic Theses and Dissertations

This dissertation deals with the evolutionary optimization of ART neural network architectures. ART (adaptive resonance theory) was introduced by a Grossberg in 1976. In the last 20 years (1987-2007) a number of ART neural network architectures were introduced into the literature (Fuzzy ARTMAP (1992), Gaussian ARTMAP (1996 and 1997) and Ellipsoidal ARTMAP (2001)). In this dissertation, we focus on the evolutionary optimization of ART neural network architectures with the intent of optimizing the size and the generalization performance of the ART neural network. A number of researchers have focused on the evolutionary optimization of neural networks, but no research has …

Pattern Recognition Via Machine Learning With Genetic Decision-Programming, Carl C. Hoff

Browse all Theses and Dissertations

In the intersection of pattern recognition, machine learning, and evolutionary computation is a new search technique by which computers might program themselves. That technique is called genetic decision-programming. A computer can gain the ability to distinguish among the things that it needs to recognize by using genetic decision-programming for pattern discovery and concept learning. Those patterns and concepts can be easily encoded in the spines of a decision program (tree or diagram). A spine consists of two parts: (1) the test-outcome pairs along a path from the program's root to any of its leaves and (2) the conclusion in that …