View All Items

Title: Prediction of survival of early stages lung cancer patients based on ER beta cellular expressions and epidemiological data.
Creator: Martinenko, Evgeny, Shivamoggi, Bhimsen, Chow, Lee, Peale, Robert, Brandenburg, John, University of Central Florida
Abstract / Description: We attempted a mathematical model for expected prognosis of lung cancer patients based ona multivariate analysis of the values of ER-interacting proteins (ERbeta) and a membranebound, glycosylated phosphoprotein MUC1), and patients clinical data recorded at the timeof initial surgery. We demonstrate that, even with the limited sample size available to use,combination of clinical and biochemical data (in particular, associated with ERbeta andMUC1) allows to predict survival of lung cancer...
Show moreWe attempted a mathematical model for expected prognosis of lung cancer patients based ona multivariate analysis of the values of ER-interacting proteins (ERbeta) and a membranebound, glycosylated phosphoprotein MUC1), and patients clinical data recorded at the timeof initial surgery. We demonstrate that, even with the limited sample size available to use,combination of clinical and biochemical data (in particular, associated with ERbeta andMUC1) allows to predict survival of lung cancer patients with about 80% accuracy whileprediction on the basis of clinical data only gives about 70% accuracy. The present work canbe viewed as a pilot study on the subject: since results conrm that ER-interacting proteinsindeed inuence lung cancer patients' survival, more data is currently being collected.
Show less
Date Issued: 2011
Identifier: CFE0004134, ucf:49120
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0004134

Title: Parallel Distributed Discrete Event Simulation Optimization Using Complexity and Deep Learning.
Creator: Cortes, Edwin, Rabelo, Luis, Lee, Gene, Kincaid, John, Elshennawy, Ahmad, University of Central Florida
Abstract / Description: Parallel distributed discrete event simulation (PDDES) is the execution of a discrete event simulation on a tightly or loosely coupled computer system with multiple processors. The discrete-event simulation model is decomposed into several logical processors (LPs) or simulation objects that can be executed concurrently using partitioning types such as spatial and temporal. PDDES is exceedingly important for the reduction of the simulation time, increase of model size, intellectual property...
Show moreParallel distributed discrete event simulation (PDDES) is the execution of a discrete event simulation on a tightly or loosely coupled computer system with multiple processors. The discrete-event simulation model is decomposed into several logical processors (LPs) or simulation objects that can be executed concurrently using partitioning types such as spatial and temporal. PDDES is exceedingly important for the reduction of the simulation time, increase of model size, intellectual property issue mitigation in multi-enterprise simulations, and the sharing of resources.One of the problems with PDDES is the time management to provide flow control over event processing, the process flow, and the coordination of different logical processors to take advantage of parallelism. Time Warp (TW), Breathing Time Buckets (BTB), and Breathing Time Warp (BTW) are three time management schemes studied by this research. For a particular PDDES problem, unfortunately, there is no clear methodology to decide a priori a time management scheme to achieve higher system and simulation performance.This dissertation shows a new approach for selecting the optimal time synchronization technique class that corresponds to a particular parallel distributed anddiscrete simulation with different levels of simulation logic complexity. Simulation complexities such as branching, parallelism, function calls, concurrency, iterations, mathematical computations, messaging frequency, event processing, and number of simulation objects interactions were given a weighted parameter value based on the cognitive weight approach. Deep belief neural networks were then used to perform deep learning from the simulation complexity parameters and their corresponding optimal time synchronization scheme value as measured by speedup performance.
Show less
Date Issued: 2015
Identifier: CFE0006211, ucf:51114
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0006211

Title: Leveraging the Intrinsic Switching Behaviors of Spintronic Devices for Digital and Neuromorphic Circuits.
Creator: Pyle, Steven, DeMara, Ronald, Vosoughi, Azadeh, Chanda, Debashis, University of Central Florida
Abstract / Description: With semiconductor technology scaling approaching atomic limits, novel approaches utilizing new memory and computation elements are sought in order to realize increased density, enhanced functionality, and new computational paradigms. Spintronic devices offer intriguing avenues to improve digital circuits by leveraging non-volatility to reduce static power dissipation and vertical integration for increased density. Novel hybrid spintronic-CMOS digital circuits are developed herein that...
Show moreWith semiconductor technology scaling approaching atomic limits, novel approaches utilizing new memory and computation elements are sought in order to realize increased density, enhanced functionality, and new computational paradigms. Spintronic devices offer intriguing avenues to improve digital circuits by leveraging non-volatility to reduce static power dissipation and vertical integration for increased density. Novel hybrid spintronic-CMOS digital circuits are developed herein that illustrate enhanced functionality at reduced static power consumption and area cost. The developed spin-CMOS D Flip-Flop offers improved power-gating strategies by achieving instant store/restore capabilities while using 10 fewer transistors than typical CMOS-only implementations. The spin-CMOS Muller C-Element developed herein improves asynchronous pipelines by reducing the area overhead while adding enhanced functionality such as instant data store/restore and delay-element-free bundled data asynchronous pipelines.Spintronic devices also provide improved scaling for neuromorphic circuits by enabling compact and low power neuron and non-volatile synapse implementations while enabling new neuromorphic paradigms leveraging the stochastic behavior of spintronic devices to realize stochastic spiking neurons, which are more akin to biological neurons and commensurate with theories from computational neuroscience and probabilistic learning rules. Spintronic-based Probabilistic Activation Function circuits are utilized herein to provide a compact and low-power neuron for Binarized Neural Networks. Two implementations of stochastic spiking neurons with alternative speed, power, and area benefits are realized. Finally, a comprehensive neuromorphic architecture comprising stochastic spiking neurons, low-precision synapses with Probabilistic Hebbian Plasticity, and a novel non-volatile homeostasis mechanism is realized for subthreshold ultra-low-power unsupervised learning with robustness to process variations. Along with several case studies, implications for future spintronic digital and neuromorphic circuits are presented.
Show less
Date Issued: 2019
Identifier: CFE0007514, ucf:52658
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0007514

Title: ATMOSPHERIC ENTRY.
Creator: Martin, Dillon A, Elgohary, Tarek, University of Central Florida
Abstract / Description: The development of atmospheric entry guidance methods is crucial to achieving the requirements for future missions to Mars; however, many missions implement a unique controller which are spacecraft specific. Here we look at the implementation of neural networks as a baseline controller that will work for a variety of different spacecraft. To accomplish this, a simulation is developed and validated with the Apollo controller. A feedforward neural network controller is then analyzed and...
Show moreThe development of atmospheric entry guidance methods is crucial to achieving the requirements for future missions to Mars; however, many missions implement a unique controller which are spacecraft specific. Here we look at the implementation of neural networks as a baseline controller that will work for a variety of different spacecraft. To accomplish this, a simulation is developed and validated with the Apollo controller. A feedforward neural network controller is then analyzed and compared to the Apollo case.
Show less
Date Issued: 2017
Identifier: CFH2000354, ucf:45874
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFH2000354

Title: GAUSS-NEWTON BASED LEARNING FOR FULLY RECURRENT NEURAL NETWORKS.
Creator: Vartak, Aniket Arun, Georgiopoulos, Michael, University of Central Florida
Abstract / Description: The thesis discusses a novel off-line and on-line learning approach for Fully Recurrent Neural Networks (FRNNs). The most popular algorithm for training FRNNs, the Real Time Recurrent Learning (RTRL) algorithm, employs the gradient descent technique for finding the optimum weight vectors in the recurrent neural network. Within the framework of the research presented, a new off-line and on-line variation of RTRL is presented, that is based on the Gauss-Newton method. The method itself is an...
Show moreThe thesis discusses a novel off-line and on-line learning approach for Fully Recurrent Neural Networks (FRNNs). The most popular algorithm for training FRNNs, the Real Time Recurrent Learning (RTRL) algorithm, employs the gradient descent technique for finding the optimum weight vectors in the recurrent neural network. Within the framework of the research presented, a new off-line and on-line variation of RTRL is presented, that is based on the Gauss-Newton method. The method itself is an approximate Newton's method tailored to the specific optimization problem, (non-linear least squares), which aims to speed up the process of FRNN training. The new approach stands as a robust and effective compromise between the original gradient-based RTRL (low computational complexity, slow convergence) and Newton-based variants of RTRL (high computational complexity, fast convergence). By gathering information over time in order to form Gauss-Newton search vectors, the new learning algorithm, GN-RTRL, is capable of converging faster to a better quality solution than the original algorithm. Experimental results reflect these qualities of GN-RTRL, as well as the fact that GN-RTRL may have in practice lower computational cost in comparison, again, to the original RTRL.
Show less
Date Issued: 2004
Identifier: CFE0000091, ucf:46065
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0000091

Title: DEVELOPMENT OF AN ARTIFICIAL NEURAL NETWORKS MODEL TO ESTIMATE DELAY USING TOLL PLAZA TRANSACTION DATA.
Creator: Muppidi, Aparna, Al-Deek, Haitham, University of Central Florida
Abstract / Description: In spite of the most up-to-date investigation of the relevant techniques to analyze the traffic characteristics and traffic operations at a toll plaza, there has not been any note worthy explorations evaluating delay from toll transaction data and using Artificial Neural Networks (ANN) at a toll plaza. This thesis lays an emphasis on the application of ANN techniques to estimate the total vehicular delay according to the lane type at a toll plaza. This is done to avoid the laborious task of...
Show moreIn spite of the most up-to-date investigation of the relevant techniques to analyze the traffic characteristics and traffic operations at a toll plaza, there has not been any note worthy explorations evaluating delay from toll transaction data and using Artificial Neural Networks (ANN) at a toll plaza. This thesis lays an emphasis on the application of ANN techniques to estimate the total vehicular delay according to the lane type at a toll plaza. This is done to avoid the laborious task of extracting data from the video recordings at a toll plaza. Based on the lane type a general methodology was developed to estimate the total vehicular delay at a toll plaza using ANN. Since there is zero delay in an Electronic Toll Collection (ETC) lane, ANN models were developed for estimating the total vehicular delay in a manual lane and automatic coin machine lane. Therefore, there are two ANN models developed in this thesis. These two ANN models were trained with three hours of data and validated with one hour of data from AM and PM peak data. The two ANN models were built with the dependent and independent variables. The dependent variables in the two models were the total vehicular delay for both the manual and automatic coin machine lane. The independent variables are those, which influence delay. A correlation analysis was performed to see if there exists any strong relationship between the dependent (outputs) and independent variables (inputs). These inputs and outputs are fed into the ANN models. The MATLABTB code was written to run the two ANN models. ANN predictions were good at estimating delay in manual lane, and delay in automatic coin machine lane.
Show less
Date Issued: 2005
Identifier: CFE0000334, ucf:46298
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0000334

Title: A NEAT APPROACH TO GENETIC PROGRAMMING.
Creator: Rodriguez, Adelein, Wu, Annie, University of Central Florida
Abstract / Description: The evolution of explicitly represented topologies such as graphs involves devising methods for mutating, comparing and combining structures in meaningful ways and identifying and maintaining the necessary topological diversity. Research has been conducted in the area of the evolution of trees in genetic programming and of neural networks and some of these problems have been addressed independently by the different research communities. In the domain of neural networks, NEAT (Neuroevolution...
Show moreThe evolution of explicitly represented topologies such as graphs involves devising methods for mutating, comparing and combining structures in meaningful ways and identifying and maintaining the necessary topological diversity. Research has been conducted in the area of the evolution of trees in genetic programming and of neural networks and some of these problems have been addressed independently by the different research communities. In the domain of neural networks, NEAT (Neuroevolution of Augmenting Topologies) has shown to be a successful method for evolving increasingly complex networks. This system's success is based on three interrelated elements: speciation, marking of historical information in topologies, and initializing search in a small structures search space. This provides the dynamics necessary for the exploration of diverse solution spaces at once and a way to discriminate between different structures. Although different representations have emerged in the area of genetic programming, the study of the tree representation has remained of interest in great part because of its mapping to programming languages and also because of the observed phenomenon of unnecessary code growth or bloat which hinders performance. The structural similarity between trees and neural networks poses an interesting question: Is it possible to apply the techniques from NEAT to the evolution of trees and if so, how does it affect performance and the dynamics of code growth? In this work we address these questions and present analogous techniques to those in NEAT for genetic programming.
Show less
Date Issued: 2007
Identifier: CFE0001971, ucf:47451
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0001971

Title: Stochastic-Based Computing with Emerging Spin-Based Device Technologies.
Creator: Bai, Yu, Lin, Mingjie, DeMara, Ronald, Wang, Jun, Jin, Yier, Dong, Yajie, University of Central Florida
Abstract / Description: In this dissertation, analog and emerging device physics is explored to provide a technology plat- form to design new bio-inspired system and novel architecture. With CMOS approaching the nano-scaling, their physics limits in feature size. Therefore, their physical device characteristics will pose severe challenges to constructing robust digital circuitry. Unlike transistor defects due to fabrication imperfection, quantum-related switching uncertainties will seriously increase their sus-...
Show moreIn this dissertation, analog and emerging device physics is explored to provide a technology plat- form to design new bio-inspired system and novel architecture. With CMOS approaching the nano-scaling, their physics limits in feature size. Therefore, their physical device characteristics will pose severe challenges to constructing robust digital circuitry. Unlike transistor defects due to fabrication imperfection, quantum-related switching uncertainties will seriously increase their sus- ceptibility to noise, thus rendering the traditional thinking and logic design techniques inadequate. Therefore, the trend of current research objectives is to create a non-Boolean high-level compu- tational model and map it directly to the unique operational properties of new, power efficient, nanoscale devices.The focus of this research is based on two-fold: 1) Investigation of the physical hysteresis switching behaviors of domain wall device. We analyze phenomenon of domain wall device and identify hys- teresis behavior with current range. We proposed the Domain-Wall-Motion-based (DWM) NCL circuit that achieves approximately 30x and 8x improvements in energy efficiency and chip layout area, respectively, over its equivalent CMOS design, while maintaining similar delay performance for a one bit full adder. 2) Investigation of the physical stochastic switching behaviors of Mag- netic Tunnel Junction (MTJ) device. With analyzing of stochastic switching behaviors of MTJ, we proposed an innovative stochastic-based architecture for implementing artificial neural network (S-ANN) with both magnetic tunneling junction (MTJ) and domain wall motion (DWM) devices, which enables efficient computing at an ultra-low voltage. For a well-known pattern recognition task, our mixed-model HSPICE simulation results have shown that a 34-neuron S-ANN imple- mentation, when compared with its deterministic-based ANN counterparts implemented with dig- ital and analog CMOS circuits, achieves more than 1.5 ? 2 orders of magnitude lower energy consumption and 2 ? 2.5 orders of magnitude less hidden layer chip area.
Show less
Date Issued: 2016
Identifier: CFE0006680, ucf:51921
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0006680

Title: D-FENS: DNS Filtering (&) Extraction Network System for Malicious Domain Names.
Creator: Spaulding, Jeffrey, Mohaisen, Aziz, Leavens, Gary, Bassiouni, Mostafa, Fu, Xinwen, Posey, Clay, University of Central Florida
Abstract / Description: While the DNS (Domain Name System) has become a cornerstone for the operation of the Internet, it has also fostered creative cases of maliciousness, including phishing, typosquatting, and botnet communication among others. To address this problem, this dissertation focuses on identifying and mitigating such malicious domain names through prior knowledge and machine learning. In the first part of this dissertation, we explore a method of registering domain names with deliberate typographical...
Show moreWhile the DNS (Domain Name System) has become a cornerstone for the operation of the Internet, it has also fostered creative cases of maliciousness, including phishing, typosquatting, and botnet communication among others. To address this problem, this dissertation focuses on identifying and mitigating such malicious domain names through prior knowledge and machine learning. In the first part of this dissertation, we explore a method of registering domain names with deliberate typographical mistakes (i.e., typosquatting) to masquerade as popular and well-established domain names. To understand the effectiveness of typosquatting, we conducted a user study which helped shed light on which techniques were more (")successful(") than others in deceiving users. While certain techniques fared better than others, they failed to take the context of the user into account. Therefore, in the second part of this dissertation we look at the possibility of an advanced attack which takes context into account when generating domain names. The main idea is determining the possibility for an adversary to improve their (")success(") rate of deceiving users with specifically-targeted malicious domain names. While these malicious domains typically target users, other types of domain names are generated by botnets for command (&) control (C2) communication. Therefore, in the third part of this dissertation we investigate domain generation algorithms (DGA) used by botnets and propose a method to identify DGA-based domain names. By analyzing DNS traffic for certain patterns of NXDomain (non-existent domain) query responses, we can accurately predict DGA-based domain names before they are registered. Given all of these approaches to malicious domain names, we ultimately propose a system called D-FENS (DNS Filtering (&) Extraction Network System). D-FENS uses machine learning and prior knowledge to accurately predict unreported malicious domain names in real-time, thereby preventing Internet devices from unknowingly connecting to a potentially malicious domain name.
Show less
Date Issued: 2018
Identifier: CFE0007587, ucf:52540
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0007587

Title: LEARNING FROM GEOMETRY IN LEARNING FOR TACTICAL AND STRATEGIC DECISION DOMAINS.
Creator: Gauci, Jason, Stanley, Kenneth, University of Central Florida
Abstract / Description: Artificial neural networks (ANNs) are an abstraction of the low-level architecture of biological brains that are often applied in general problem solving and function approximation. Neuroevolution (NE), i.e. the evolution of ANNs, has proven effective at solving problems in a variety of domains. Information from the domain is input to the ANN, which outputs its desired actions. This dissertation presents a new NE algorithm called Hypercube-based NeuroEvolution of Augmenting Topologies ...
Show moreArtificial neural networks (ANNs) are an abstraction of the low-level architecture of biological brains that are often applied in general problem solving and function approximation. Neuroevolution (NE), i.e. the evolution of ANNs, has proven effective at solving problems in a variety of domains. Information from the domain is input to the ANN, which outputs its desired actions. This dissertation presents a new NE algorithm called Hypercube-based NeuroEvolution of Augmenting Topologies (HyperNEAT), based on a novel indirect encoding of ANNs. The key insight in HyperNEAT is to make the algorithm aware of the geometry in which the ANNs are embedded and thereby exploit such domain geometry to evolve ANNs more effectively. The dissertation focuses on applying HyperNEAT to tactical and strategic decision domains. These domains involve simultaneously considering short-term tactics while also balancing long-term strategies. Board games such as checkers and Go are canonical examples of such domains; however, they also include real-time strategy games and military scenarios. The dissertation details three proposed extensions to HyperNEAT designed to work in tactical and strategic decision domains. The first is an action selector ANN architecture that allows the ANN to indicate its judgements on every possible action all at once. The second technique is called substrate extrapolation. It allows learning basic concepts at a low resolution, and then increasing the resolution to learn more advanced concepts. The final extension is geometric game-tree pruning, whereby HyperNEAT can endow the ANN the ability to focus on specific areas of a domain (such as a checkers board) that deserve more inspection. The culminating contribution is to demonstrate the ability of HyperNEAT with these extensions to play Go, a most challenging game for artificial intelligence, by combining HyperNEAT with UCT.
Show less
Date Issued: 2010
Identifier: CFE0003464, ucf:48962
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0003464

Title: Online Neuro-Adaptive Learning For Power System Dynamic State Estimation.
Creator: Birari, Rahul, Zhou, Qun, Sun, Wei, Dimitrovski, Aleksandar, University of Central Florida
Abstract / Description: With the increased penetration of renewable generation in the smart grid , it is crucial to have knowledge of rapid changes of system states. The information of real-time electro-mechanical dynamic states of generators are essential to ensuring reliability and detecting instability of the grid. The conventional SCADA based Dynamic State Estimation (DSE) was limited by the slow sampling rates (2-4 Hz). With the advent of PMU based synchro-phasor technology in tandem with Wide Area Monitoring...
Show moreWith the increased penetration of renewable generation in the smart grid , it is crucial to have knowledge of rapid changes of system states. The information of real-time electro-mechanical dynamic states of generators are essential to ensuring reliability and detecting instability of the grid. The conventional SCADA based Dynamic State Estimation (DSE) was limited by the slow sampling rates (2-4 Hz). With the advent of PMU based synchro-phasor technology in tandem with Wide Area Monitoring System (WAMS), it has become possible to avail rapid real-time measurements at the network nodes. These measurements can be exploited for better estimates of system dynamic states. In this research, we have proposed a novel Artificial Intelligence (AI) based real-time neuro-adaptive algorithm for rotor angle and speed estimation of synchronous generators. Generator swing equations and power flow models are incorporated in the online learning. The algorithm learns and adapts in real-time to achieve accurate estimates. Simulation is carried out on 68 bus 16 generator NETS-NYPS model. The neuro-adaptive algorithm is compared with classical Kalman Filter based DSE. Applicability and accuracy of the proposed method is demonstrated under the influence of noise and faulty conditions.
Show less
Date Issued: 2017
Identifier: CFE0006858, ucf:51747
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0006858

Title: NEURAL NETWORK TREES AND SIMULATION DATABASES: NEW APPROACHES FOR SIGNALIZED INTERSECTION CRASH CLASSIFICATION AND PREDICTION.
Creator: Nawathe, Piyush, Abdel-Aty, Mohamed, University of Central Florida
Abstract / Description: Intersection related crashes form a significant proportion of the crashes occurring on roadways. Many organizations such as the Federal Highway Administration (FHWA) and American Association of State Highway and Transportation Officials (AASHTO) are considering intersection safety improvement as one of their top priority areas. This study contributes to the area of safety of signalized intersections by identifying the traffic and geometric characteristics that affect the different types of...
Show moreIntersection related crashes form a significant proportion of the crashes occurring on roadways. Many organizations such as the Federal Highway Administration (FHWA) and American Association of State Highway and Transportation Officials (AASHTO) are considering intersection safety improvement as one of their top priority areas. This study contributes to the area of safety of signalized intersections by identifying the traffic and geometric characteristics that affect the different types of crashes. The first phase of this thesis was to classify the crashes occurring at signalized intersections into rear-end, angle, turn and sideswipe crash types based on the traffic and geometric properties of the intersections and the conditions at the time of the crashes. This was achieved by using an innovative approach developed in this thesis "Neural Network Trees". The first neural network model built in the Neural Network tree classified the crashes either into rear end and sideswipe or into angle and turn crashes. The next models further classified the crashes into their individual types. Two different neural network methods (MLP and PNN) were used in classification, and the neural network with a better performance was selected for each model. For these models, the significant variables were identified using the forward sequential selection method. Then a large simulation database was built that contained all possible combinations of intersections subjected to various crash conditions. The collision type of crashes was predicted for this simulation database and the output obtained was plotted along with the input variables to obtain a relationship between the input and output variables. For example, the analysis showed that the number of rear end and sideswipe crashes increase relative to the angle and turn crashes when there is an increase in the major and minor roadways' AADT and speed limits, surface conditions, total left turning lanes, channelized right turning lanes for the major roadway and the protected left turning lanes for the minor roadway, but decrease when the light conditions are dark. The next phase in this study was to predict the frequency of different types of crashes at signalized intersections by using the geometric and traffic characteristics of the intersections. A high accuracy in predicting the crash frequencies was obtained by using another innovative method where the intersections were first classified into two different types named the "safe" and "unsafe" intersections based on the total number of lanes at the intersections and then the frequency of crashes was predicted for each type of intersections separately. This method consisted of identifying the best neural network for each step of the analysis, selecting significant variables, using a different simulation database that contained all possible combinations of intersections and then plotting each input variable with the average output to obtain the pattern in which the frequency of crashes will vary based on the changes in the geometric and traffic characteristics of the intersections. The patterns indicated that an increase in the number of lanes of the major roadway, lanes of the minor roadway and the AADT on the major roadway leads to an increased crashes of all types, whereas an increase in protected left turning lanes on the major road increases the rear end and sideswipe crashes but decreases the angle, turning and overall crash frequencies. The analyses performed in this thesis were possible due to a diligent data collection effort. Traffic and geometric characteristics were obtained from multiple sources for 1562 signalized intersections in Brevard, Hillsborough, Miami-Dade, Seminole and Orange counties and the city of Orlando in Florida. The crash database for these intersections contained 27,044 crashes. This research sheds a light on the characteristics of different types of crashes. The method used in classifying crashes into their respective collision types provides a deeper insight on the characteristics of each type of crash and can be helpful in mitigating a particular type of crash at an intersection. The second analysis carried out has a three fold advantage. First, it identifies if an intersection can be considered safe for different crash types. Second, it accurately predicts the frequencies of total, rear end, angle, sideswipe and turn crashes. Lastly, it identifies the traffic and geometric characteristics of signalized intersections that affect each of these crash types. Thus the models developed in this thesis can be used to identify the specific problems at an intersection, and identify the factors that should be changed to improve its safety
Show less
Date Issued: 2005
Identifier: CFE0000664, ucf:46524
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0000664

Title: NEURAL NETWORKS SATISFYING STONE-WEIESTRASS THEOREM AND APPROXIMATING SCATTERED DATABYKOHONEN NEURAL NETWORKS.
Creator: Thakkar, Pinal, Mohapatra, Ram, University of Central Florida
Abstract / Description: Neural networks are an attempt to build computer networks called artificial neurons, which imitate the activities of the human brain. Its origin dates back to 1943 when neurophysiologist Warren Me Cello and logician Walter Pits produced the first artificial neuron. Since then there has been tremendous development of neural networks and their applications to pattern and optical character recognition, speech processing, time series prediction, image processing and scattered data approximation....
Show moreNeural networks are an attempt to build computer networks called artificial neurons, which imitate the activities of the human brain. Its origin dates back to 1943 when neurophysiologist Warren Me Cello and logician Walter Pits produced the first artificial neuron. Since then there has been tremendous development of neural networks and their applications to pattern and optical character recognition, speech processing, time series prediction, image processing and scattered data approximation. Since it has been shown that neural nets can approximate all but pathological functions, Neil Cotter considered neural network architecture based on Stone-Weierstrass Theorem. Using exponential functions, polynomials, rational functions and Boolean functions one can follow the method given by Cotter to obtain neural networks, which can approximate bounded measurable functions. Another problem of current research in computer graphics is to construct curves and surfaces from scattered spatial points by using B-Splines and NURBS or Bezier surfaces. Hoffman and Varady used Kohonen neural networks to construct appropriate grids. This thesis is concerned with two types of neural networks viz. those which satisfy the conditions of the Stone-Weierstrass theorem and Kohonen neural networks. We have used self-organizing maps for scattered data approximation. Neural network Tool Box from MATLAB is used to develop the required grids for approximating scattered data in one and two dimensions.
Show less
Date Issued: 2004
Identifier: CFE0000226, ucf:46262
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0000226

Title: MODIFICATIONS TO THE FUZZY-ARTMAP ALGORITHM FOR DISTRIBUTED LEARNING IN LARGE DATA SETS.
Creator: Castro, Jose R, Georgiopoulos, Michael, University of Central Florida
Abstract / Description: The Fuzzy-ARTMAP (FAM) algorithm is one of the premier neural network architectures for classification problems. FAM can learn on line and is usually faster than other neural network approaches. Nevertheless the learning time of FAM can slow down considerably when the size of the training set increases into the hundreds of thousands. We apply data partitioning and networkpartitioning to the FAM algorithm in a sequential and parallel settingto achieve better convergence time and to efficiently...
Show moreThe Fuzzy-ARTMAP (FAM) algorithm is one of the premier neural network architectures for classification problems. FAM can learn on line and is usually faster than other neural network approaches. Nevertheless the learning time of FAM can slow down considerably when the size of the training set increases into the hundreds of thousands. We apply data partitioning and networkpartitioning to the FAM algorithm in a sequential and parallel settingto achieve better convergence time and to efficiently train withlarge databases (hundreds of thousands of patterns).Our parallelization is implemented on a Beowulf clusters of workstations. Two data partitioning approaches and two networkpartitioning approaches are developed. Extensive testing of all the approaches is done on three large datasets (half a milliondata points). One of them is the Forest Covertype database from Blackard and the other two are artificially generated Gaussian data with different percentages of overlap between classes.Speedups in the data partitioning approach reached the order of the hundreds without having to invest in parallel computation. Speedups onthe network partitioning approach are close to linear on a cluster of workstations. Both methods allowed us to reduce the computation time of training the neural network in large databases from days to minutes. We prove formally that the workload balance of our network partitioning approaches will never be worse than an acceptable bound, and also demonstrate the correctness of these parallelization variants of FAM.
Show less
Date Issued: 2004
Identifier: CFE0000065, ucf:46092
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0000065

Title: EVOLUTIONARY OPTIMIZATION OF SUPPORT VECTOR MACHINES.
Creator: Gruber, Fred, Rabelo, Luis, University of Central Florida
Abstract / Description: Support vector machines are a relatively new approach for creating classifiers that have become increasingly popular in the machine learning community. They present several advantages over other methods like neural networks in areas like training speed, convergence, complexity control of the classifier, as well as a stronger mathematical background based on optimization and statistical learning theory. This thesis deals with the problem of model selection with support vector machines, that is...
Show moreSupport vector machines are a relatively new approach for creating classifiers that have become increasingly popular in the machine learning community. They present several advantages over other methods like neural networks in areas like training speed, convergence, complexity control of the classifier, as well as a stronger mathematical background based on optimization and statistical learning theory. This thesis deals with the problem of model selection with support vector machines, that is, the problem of finding the optimal parameters that will improve the performance of the algorithm. It is shown that genetic algorithms provide an effective way to find the optimal parameters for support vector machines. The proposed algorithm is compared with a backpropagation Neural Network in a dataset that represents individual models for electronic commerce.
Show less
Date Issued: 2004
Identifier: CFE0000244, ucf:46251
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0000244

Title: NON-SILICON MICROFABRICATED NANOSTRUCTURED CHEMICAL SENSORS FOR ELECTRIC NOSE APPLICATION.
Creator: Gong, Jianwei, Chen, Quanfang, University of Central Florida
Abstract / Description: A systematic investigation has been performed for "Electric Nose", a system that can identify gas samples and detect their concentrations by combining sensor array and data processing technologies. Non-silicon based microfabricatition has been developed for micro-electro-mechanical-system (MEMS) based gas sensors. Novel sensors have been designed, fabricated and tested. Nanocrystalline semiconductor metal oxide (SMO) materials include SnO2, WO3 and In2O3 have been studied for gas sensing...
Show moreA systematic investigation has been performed for "Electric Nose", a system that can identify gas samples and detect their concentrations by combining sensor array and data processing technologies. Non-silicon based microfabricatition has been developed for micro-electro-mechanical-system (MEMS) based gas sensors. Novel sensors have been designed, fabricated and tested. Nanocrystalline semiconductor metal oxide (SMO) materials include SnO2, WO3 and In2O3 have been studied for gas sensing applications. Different doping material such as copper, silver, platinum and indium are studied in order to achieve better selectivity for different targeting toxic gases including hydrogen, carbon monoxide, hydrogen sulfide etc. Fundamental issues like sensitivity, selectivity, stability, temperature influence, humidity influence, thermal characterization, drifting problem etc. of SMO gas sensors have been intensively investigated. A novel approach to improve temperature stability of SMO (including tin oxide) gas sensors by applying a temperature feedback control circuit has been developed. The feedback temperature controller that is compatible with MEMS sensor fabrication has been invented and applied to gas sensor array system. Significant improvement of stability has been achieved compared to SMO gas sensors without temperature compensation under the same ambient conditions. Single walled carbon nanotube (SWNT) has been studied to improve SnO2 gas sensing property in terms of sensitivity, response time and recovery time. Three times of better sensitivity has been achieved experimentally. The feasibility of using TSK Fuzzy neural network algorithm for Electric Nose has been exploited during the research. A training process of using TSK Fuzzy neural network with input/output pairs from individual gas sensor cell has been developed. This will make electric nose smart enough to measure gas concentrations in a gas mixture. The model has been proven valid by gas experimental results conducted.
Show less
Date Issued: 2005
Identifier: CFE0000377, ucf:46328
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0000377

Title: LEARNING HUMAN BEHAVIOR FROM OBSERVATION FOR GAMING APPLICATIONS.
Creator: Moriarty, Christopher, Gonzalez, Avelino, University of Central Florida
Abstract / Description: The gaming industry has reached a point where improving graphics has only a small effect on how much a player will enjoy a game. One focus has turned to adding more humanlike characteristics into computer game agents. Machine learning techniques are being used scarcely in games, although they do offer powerful means for creating humanlike behaviors in agents. The first person shooter (FPS), Quake 2, is an open source game that offers a multi-agent environment to create game agents (bots) in....
Show moreThe gaming industry has reached a point where improving graphics has only a small effect on how much a player will enjoy a game. One focus has turned to adding more humanlike characteristics into computer game agents. Machine learning techniques are being used scarcely in games, although they do offer powerful means for creating humanlike behaviors in agents. The first person shooter (FPS), Quake 2, is an open source game that offers a multi-agent environment to create game agents (bots) in. This work attempts to combine neural networks with a modeling paradigm known as context based reasoning (CxBR) to create a contextual game observation (CONGO) system that produces Quake 2 agents that behave as a human player trains them to act. A default level of intelligence is instilled into the bots through contextual scripts to prevent the bot from being trained to be completely useless. The results show that the humanness and entertainment value as compared to a traditional scripted bot have improved, although, CONGO bots usually ranked only slightly above a novice skill level. Overall, CONGO is a technique that offers the gaming community a mode of game play that has promising entertainment value.
Show less
Date Issued: 2007
Identifier: CFE0001694, ucf:47201
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0001694

Title: Leaning Robust Sequence Features via Dynamic Temporal Pattern Discovery.
Creator: Hu, Hao, Wang, Liqiang, Zhang, Shaojie, Liu, Fei, Qi, GuoJun, Zhou, Qun, University of Central Florida
Abstract / Description: As a major type of data, time series possess invaluable latent knowledge for describing the real world and human society. In order to improve the ability of intelligent systems for understanding the world and people, it is critical to design sophisticated machine learning algorithms for extracting robust time series features from such latent knowledge. Motivated by the successful applications of deep learning in computer vision, more and more machine learning researchers put their attentions...
Show moreAs a major type of data, time series possess invaluable latent knowledge for describing the real world and human society. In order to improve the ability of intelligent systems for understanding the world and people, it is critical to design sophisticated machine learning algorithms for extracting robust time series features from such latent knowledge. Motivated by the successful applications of deep learning in computer vision, more and more machine learning researchers put their attentions on the topic of applying deep learning techniques to time series data. However, directly employing current deep models in most time series domains could be problematic. A major reason is that temporal pattern types that current deep models are aiming at are very limited, which cannot meet the requirement of modeling different underlying patterns of data coming from various sources. In this study we address this problem by designing different network structures explicitly based on specific domain knowledge such that we can extract features via most salient temporal patterns. More specifically, we mainly focus on two types of temporal patterns: order patterns and frequency patterns. For order patterns, which are usually related to brain and human activities, we design a hashing-based neural network layer to globally encode the ordinal pattern information into the resultant features. It is further generalized into a specially designed Recurrent Neural Networks (RNN) cell which can learn order patterns in an online fashion. On the other hand, we believe audio-related data such as music and speech can benefit from modeling frequency patterns. Thus, we do so by developing two types of RNN cells. The first type tries to directly learn the long-term dependencies on frequency domain rather than time domain. The second one aims to dynamically filter out the ``noise" frequencies based on temporal contexts. By proposing various deep models based on different domain knowledge and evaluating them on extensive time series tasks, we hope this work can provide inspirations for others and increase the community's interests on the problem of applying deep learning techniques to more time series tasks.
Show less
Date Issued: 2019
Identifier: CFE0007470, ucf:52679
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0007470

Title: Training Neural Networks Through the Integration of Evolution and Gradient Descent.
Creator: Morse, Gregory, Stanley, Kenneth, Wu, Annie, Shah, Mubarak, Wiegand, Rudolf, University of Central Florida
Abstract / Description: Neural networks have achieved widespread adoption due to both their applicability to a wide range of problems and their success relative to other machine learning algorithms. The training of neural networks is achieved through any of several paradigms, most prominently gradient-based approaches (including deep learning), but also through up-and-coming approaches like neuroevolution. However, while both of these neural network training paradigms have seen major improvements over the past...
Show moreNeural networks have achieved widespread adoption due to both their applicability to a wide range of problems and their success relative to other machine learning algorithms. The training of neural networks is achieved through any of several paradigms, most prominently gradient-based approaches (including deep learning), but also through up-and-coming approaches like neuroevolution. However, while both of these neural network training paradigms have seen major improvements over the past decade, little work has been invested in developing algorithms that incorporate the advances from both deep learning and neuroevolution. This dissertation introduces two new algorithms that are steps towards the integration of gradient descent and neuroevolution for training neural networks. The first is (1) the Limited Evaluation Evolutionary Algorithm (LEEA), which implements a novel form of evolution where individuals are partially evaluated, allowing rapid learning and enabling the evolutionary algorithm to behave more like gradient descent. This conception provides a critical stepping stone to future algorithms that more tightly couple evolutionary and gradient descent components. The second major algorithm (2) is Divergent Discriminative Feature Accumulation (DDFA), which combines a neuroevolution phase, where features are collected in an unsupervised manner, with a gradient descent phase for fine tuning of the neural network weights. The neuroevolution phase of DDFA utilizes an indirect encoding and novelty search, which are sophisticated neuroevolution components rarely incorporated into gradient descent-based systems. Further contributions of this work that build on DDFA include (3) an empirical analysis to identify an effective distance function for novelty search in high dimensions and (4) the extension of DDFA for the purpose of discovering convolutional features. The results of these DDFA experiments together show that DDFA discovers features that are effective as a starting point for gradient descent, with significant improvement over gradient descent alone. Additionally, the method of collecting features in an unsupervised manner allows DDFA to be applied to domains with abundant unlabeled data and relatively sparse labeled data. This ability is highlighted in the STL-10 domain, where DDFA is shown to make effective use of unlabeled data.
Show less
Date Issued: 2019
Identifier: CFE0007840, ucf:52819
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0007840

Title: detecting anomalies from big data system logs.
Creator: Lu, Siyang, Wang, Liqiang, Zhang, Shaojie, Zhang, Wei, Wu, Dazhong, University of Central Florida
Abstract / Description: Nowadays, big data systems (e.g., Hadoop and Spark) are being widely adopted by many domains for offering effective data solutions, such as manufacturing, healthcare, education, and media. A common problem about big data systems is called anomaly, e.g., a status deviated from normal execution, which decreases the performance of computation or kills running programs. It is becoming a necessity to detect anomalies and analyze their causes. An effective and economical approach is to analyze...
Show moreNowadays, big data systems (e.g., Hadoop and Spark) are being widely adopted by many domains for offering effective data solutions, such as manufacturing, healthcare, education, and media. A common problem about big data systems is called anomaly, e.g., a status deviated from normal execution, which decreases the performance of computation or kills running programs. It is becoming a necessity to detect anomalies and analyze their causes. An effective and economical approach is to analyze system logs. Big data systems produce numerous unstructured logs that contain buried valuable information. However manually detecting anomalies from system logs is a tedious and daunting task.This dissertation proposes four approaches that can accurately and automatically analyze anomalies from big data system logs without extra monitoring overhead. Moreover, to detect abnormal tasks in Spark logs and analyze root causes, we design a utility to conduct fault injection and collect logs from multiple compute nodes. (1) Our first method is a statistical-based approach that can locate those abnormal tasks and calculate the weights of factors for analyzing the root causes. In the experiment, four potential root causes are considered, i.e., CPU, memory, network, and disk I/O. The experimental results show that the proposed approach is accurate in detecting abnormal tasks as well as finding the root causes. (2) To give a more reasonable probability result and avoid ad-hoc factor weights calculating, we propose a neural network approach to analyze root causes of abnormal tasks. We leverage General Regression Neural Network (GRNN) to identify root causes for abnormal tasks. The likelihood of reported root causes is presented to users according to the weighted factors by GRNN. (3) To further improve anomaly detection by avoiding feature extraction, we propose a novel approach by leveraging Convolutional Neural Networks (CNN). Our proposed model can automatically learn event relationships in system logs and detect anomaly with high accuracy. Our deep neural network consists of logkey2vec embeddings, three 1D convolutional layers, a dropout layer, and max pooling. According to our experiment, our CNN-based approach has better accuracy compared to other approaches using Long Short-Term Memory (LSTM) and Multilayer Perceptron (MLP) on detecting anomaly in Hadoop DistributedFile System (HDFS) logs. (4) To analyze system logs more accurately, we extend our CNN-based approach with two attention schemes to detect anomalies in system logs. The proposed two attention schemes focus on different features from CNN's output. We evaluate our approaches with several benchmarks, and the attention-based CNN model shows the best performance among all state-of-the-art methods.
Show less
Date Issued: 2019
Identifier: CFE0007673, ucf:52499
Format: Document (PDF)
PURL: http://purl.flvc.org/ucf/fd/CFE0007673

neural networks (10)	+ -
Neural Networks (7)	+ -
deep learning (4)	+ -
machine learning (4)	+ -
Machine Learning (3)	+ -

NEAT (3)	+ -
Neural Network (3)	+ -
Artificial Intelligence (2)	+ -
Artificial Neural Networks (2)	+ -
Convolutional Neural Network (2)	+ -
Genetic Algorithms (2)	+ -
HyperNEAT (2)	+ -
Neuroevolution (2)	+ -
Stochastic Computing (2)	+ -
artificial neural network (2)	+ -
convolutional neural network (2)	+ -
evolutionary computation (2)	+ -
genetic algorithms (2)	+ -
genetic programming (2)	+ -
neural network (2)	+ -
remote sensing (2)	+ -
APPROXIMATION OF SCATTERED DATA (1)	+ -
Academic (1)	+ -
Adaptive neuro fuzzy inference systems (1)	+ -
Adverse events (1)	+ -
Anomaly detection (1)	+ -
Artificial Evolution (1)	+ -
Artificial neural network (1)	+ -
Artificial neural networks (1)	+ -
Atmopheric Entry (1)	+ -

ETD Collection (41)	+ -
Honors Collection (1)	+ -

University of Central Florida (41)	+ -
Stanley, Kenneth (6)	+ -
Georgiopoulos, Michael (5)	+ -
Sukthankar, Gita (5)	+ -
Wu, Annie (4)	+ -

DeMara, Ronald (3)	+ -
Gonzalez, Avelino (3)	+ -
Shah, Mubarak (3)	+ -
Zhang, Shaojie (3)	+ -
Zhou, Qun (3)	+ -
Chang, Ni-bin (2)	+ -
Dimitrovski, Aleksandar (2)	+ -
Elshennawy, Ahmad (2)	+ -
Foroosh, Hassan (2)	+ -
Garibay, Ivan (2)	+ -
Lee, Gene (2)	+ -
Lin, Mingjie (2)	+ -
Mikhael, Wasfy (2)	+ -
Mohapatra, Ram (2)	+ -
Qi, GuoJun (2)	+ -
Rabelo, Luis (2)	+ -
Wang, Jun (2)	+ -
Wang, Liqiang (2)	+ -
Wanielista, Martin (2)	+ -
Wiegand, Rudolf (2)	+ -
Zhang, Wei (2)	+ -
Abdel-Aty, Mohamed (1)	+ -
Al-Daraiseh, Ahmad (1)	+ -
Al-Deek, Haitham (1)	+ -
Al-Haj Hussein, Ala (1)	+ -

Search In

Pages

Pages

Sort Results By:

Narrow results by:

Topical Subject

Collection

Type of Resource

Language

Creator

Format

Owner