Digital Commons Network^™

Empirical Methods For Predicting Student Retention- A Summary From The Literature, Matt Bogard

Economics Faculty Publications

The vast majority of the literature related to the empirical estimation of retention models includes a discussion of the theoretical retention framework established by Bean, Braxton, Tinto, Pascarella, Terenzini and others (see Bean, 1980; Bean, 2000; Braxton, 2000; Braxton et al, 2004; Chapman and Pascarella, 1983; Pascarell and Ternzini, 1978; St. John and Cabrera, 2000; Tinto, 1975) This body of research provides a starting point for the consideration of which explanatory variables to include in any model specification, as well as identifying possible data sources. The literature separates itself into two major camps including research related to the hypothesis testing …

Empirical Methods-A Review: With An Introduction To Data Mining And Machine Learning, Matt Bogard

Economics Faculty Publications

This presentation was part of a staff workshop focused on empirical methods and applied research. This includes a basic overview of regression with matrix algebra, maximum likelihood, inference, and model assumptions. Distinctions are made between paradigms related to classical statistical methods and algorithmic approaches. The presentation concludes with a brief discussion of generalization error, data partitioning, decision trees, and neural networks.

Narrative Analysis And Computational Model To Predict Interestingness Of Narratives, Laxman Thapa

Theses and Dissertations - UTB/UTPA

In this research, I present results demonstrating the classification of the specially generated narratives by a machine agent by listening to human subject describing the same sets of the events. These classifications are based on human ratings of interestingness for many different recountings of the same stories. The classification is performed on various features selected after analyzing the different possible feature that affect on the interestingness of narratives. The features were extracted from the surface text as well as from annotations of how each narration relates to the content of the known story. I present the annotation process and resulting …

Learning Local Features Using Boosted Trees For Face Recognition, Rajkiran Gottumukkal

Electrical & Computer Engineering Theses & Dissertations

Face recognition is fundamental to a number of significant applications that include but not limited to video surveillance and content based image retrieval. Some of the challenges which make this task difficult are variations in faces due to changes in pose, illumination and deformation. This dissertation proposes a face recognition system to overcome these difficulties. We propose methods for different stages of face recognition which will make the system more robust to these variations. We propose a novel method to perform skin segmentation which is fast and able to perform well under different illumination conditions. We also propose a method …

On The Effect Of Criticality And Topology On Learning In Random Boolean Networks, Alireza Goudarzi

Systems Science Friday Noon Seminar Series

Random Boolean networks (RBN) are discrete dynamical systems composed of N automata with a binary state, each of which interacts with other automata in the network. RBNs were originally introduced as simplified models of gene regulation. In this presentation, I will present recent work done conjointly with Natali Gulbahce (UCSF), Thimo Rohlf (MPI, CNRS), and Christof Teuscher (PSU). We extend the study of learning in feedforward Boolean networks to random Boolean networks (RBNs) and systematically explore the relationship between the learning capability, the network topology, the system size N, the training sample T, and the complexity of the computational task. …

Data Mining Based Learning Algorithms For Semi-Supervised Object Identification And Tracking, Michael P. Dessauer

Doctoral Dissertations

Sensor exploitation (SE) is the crucial step in surveillance applications such as airport security and search and rescue operations. It allows localization and identification of movement in urban settings and can significantly boost knowledge gathering, interpretation and action. Data mining techniques offer the promise of precise and accurate knowledge acquisition techniques in high-dimensional data domains (and diminishing the “curse of dimensionality” prevalent in such datasets), coupled by algorithmic design in feature extraction, discriminative ranking, feature fusion and supervised learning (classification). Consequently, data mining techniques and algorithms can be used to refine and process captured data and to detect, recognize, classify, …

An Exploration Of Multi-Agent Learning Within The Game Of Sheephead, Brady Brau

All Graduate Theses, Dissertations, and Other Capstone Projects

In this paper, we examine a machine learning technique presented by Ishii et al. used to allow for learning in a multi-agent environment and apply an adaptation of this learning technique to the card game Sheephead. We then evaluate the effectiveness of our adaptation by running simulations against rule-based opponents. Multi-agent learning presents several layers of complexity on top of a single-agent learning in a stationary environment. This added complexity and increased state space is just beginning to be addressed by researchers. We utilize techniques used by Ishii et al. to facilitate this multi-agent learning. We model the environment of …

Effective Task Transfer Through Indirect Encoding, Phillip Verbancsics

Electronic Theses and Dissertations

An important goal for machine learning is to transfer knowledge between tasks. For example, learning to play RoboCup Keepaway should contribute to learning the full game of RoboCup soccer. Often approaches to task transfer focus on transforming the original representation to fit the new task. Such representational transformations are necessary because the target task often requires new state information that was not included in the original representation. In RoboCup Keepaway, changing from the 3 vs. 2 variant of the task to 4 vs. 3 adds state information for each of the new players. In contrast, this dissertation explores the idea …

Optimal Detour Planning Around Blocked Construction Zones, Mutasem Jardaneh

Electronic Theses and Dissertations

Construction zones are traffic way areas where construction, maintenance or utility work is identified by warning signs, signals and indicators, including those on transport devices that mark the beginning and end of construction zones. Construction zones are among the most dangerous work areas, with workers facing workplace safety challenges that often lead to catastrophic injuries or fatalities. In addition, daily commuters are also impacted by construction zone detours that affect their safety and daily commute time. These problems represent major challenges to construction planners as they are required to plan vehicle routes around construction zones in such a way that …

Prediction Of Brain Tumor Progression Using Multiple Histogram Matched Mri Scans, Debrup Banerjee, Loc Tran, Jiang Li, Yuzhong Shen, Frederic Mckenzie, Jihong Wang, Ronald M. Summers (Ed.), Bram Van Ginneken (Ed.)

Electrical & Computer Engineering Faculty Publications

In a recent study [1], we investigated the feasibility of predicting brain tumor progression based on multiple MRI series and we tested our methods on seven patients' MRI images scanned at three consecutive visits A, B and C. Experimental results showed that it is feasible to predict tumor progression from visit A to visit C using a model trained by the information from visit A to visit B. However, the trained model failed when we tried to predict tumor progression from visit B to visit C, though it is clinically more important. Upon a closer look at the MRI scans …

Energy-Economical Heuristically Based Control Of Compass Gait Walking On Stochastically Varying Terrain, Christian Hubicki

Master’s Theses

Investigation uses simulation to explore the inherent tradeoffs ofcontrolling high-speed and highly robust walking robots while minimizing energy consumption. Using a novel controller which optimizes robustness, energy economy, and speed of a simulated robot on rough terrain, the user can adjust their priorities between these three outcome measures and systematically generate a performance curveassessing the tradeoffs associated with these metrics.

Assessing Data Quality In A Sensor Network For Environmental Monitoring, Gesuri Ramirez

Open Access Theses & Dissertations

Assessing the quality of sensor data in environmental monitoring applications is important, as erroneous readings produced by malfunctioning sensors, calibration drift, and problematic climatic conditions, such as icing or dust, are common.Traditional data quality checking and correction is a painstaking manual process, so the development of automatic systems for this task is highly desirable.

This study investigates machine learning methods to identify and clean incorrect data from a real-world environmental sensor network, the Jornada Experimental Range, located in Southern New Mexico. We evaluated several learning algorithms and data replacement schemes, and developed a method to identify the problematic sensor. The …

Algorithms For Training Large-Scale Linear Programming Support Vector Regression And Classification, Pablo Rivas Perea

Open Access Theses & Dissertations

The main contribution of this dissertation is the development of a method to train a Support Vector Regression (SVR) model for the large-scale case where the number of training samples supersedes the computational resources. The proposed scheme consists of posing the SVR problem entirely as a Linear Programming (LP) problem and on the development of a sequential optimization method based on variables decomposition, constraints decomposition, and the use of primal-dual interior point methods. Experimental results demonstrate that the proposed approach has comparable performance with other SV-based classifiers. Particularly, experiments demonstrate that as the problem size increases, the sparser the solution …

Prestructuring Multilayer Perceptrons Based On Information-Theoretic Modeling Of A Partido-Alto-Based Grammar For Afro-Brazilian Music: Enhanced Generalization And Principles Of Parsimony, Including An Investigation Of Statistical Paradigms, Mehmet Vurkaç

Dissertations and Theses

The present study shows that prestructuring based on domain knowledge leads to statistically significant generalization-performance improvement in artificial neural networks (NNs) of the multilayer perceptron (MLP) type, specifically in the case of a noisy real-world problem with numerous interacting variables.

The prestructuring of MLPs based on knowledge of the structure of a problem domain has previously been shown to improve generalization performance. However, the problem domains for those demonstrations suffered from significant shortcomings: 1) They were purely logical problems, and 2) they contained small numbers of variables in comparison to most data-mining applications today. Two implications of the former were …

Optimal Control For Autonomous Motor Behavior, Tom Erez

All Theses and Dissertations (ETDs)

This dissertation presents algorithms that allow robots to generate optimal behavior from first principles. Instead of hard-coding every desired behavior, we encode the task as a cost function, and use numerical optimization to find action sequences that can accomplish the task. Using the theoretical framework of optimal control, we develop methods for generating autonomous motor behavior in high-dimensional domains of legged locomotion. We identify three foundational problems that limit the application of existing optimal control algorithms, and present guiding principles that address these issues. First, some traditional algorithms use global optimization, where every possible state is considered. This approach cannot …

Combining Natural Language Processing And Statistical Text Mining: A Study Of Specialized Versus Common Languages, Jay Jarman

USF Tampa Graduate Theses and Dissertations

This dissertation focuses on developing and evaluating hybrid approaches for analyzing free-form text in the medical domain. This research draws on natural language processing (NLP) techniques that are used to parse and extract concepts based on a controlled vocabulary. Once important concepts are extracted, additional machine learning algorithms, such as association rule mining and decision tree induction, are used to discover classification rules for specific targets. This multi-stage pipeline approach is contrasted with traditional statistical text mining (STM) methods based on term counts and term-by-document frequencies. The aim is to create effective text analytic processes by adapting and combining individual …

Hierarchical Temporal Memory Cortical Learning Algorithm For Pattern Recognition On Multi-Core Architectures, Ryan William Price

Dissertations and Theses

Strongly inspired by an understanding of mammalian cortical structure and function, the Hierarchical Temporal Memory Cortical Learning Algorithm (HTM CLA) is a promising new approach to problems of recognition and inference in space and time. Only a subset of the theoretical framework of this algorithm has been studied, but it is already clear that there is a need for more information about the performance of HTM CLA with real data and the associated computational costs. For the work presented here, a complete implementation of Numenta's current algorithm was done in C++. In validating the implementation, first and higher order sequence …

Features Extraction Using Random Matrix Theory., Viktoria Borisovna Rojkova

Electronic Theses and Dissertations

Representing the complex data in a concise and accurate way is a special stage in data mining methodology. Redundant and noisy data affects generalization power of any classification algorithm, undermines the results of any clustering algorithm and finally encumbers the monitoring of large dynamic systems. This work provides several efficient approaches to all aforementioned sides of the analysis. We established, that notable difference can be made, if the results from the theory of ensembles of random matrices are employed. Particularly important result of our study is a discovered family of methods based on projecting the data set on different subsets …

Identification And Optimization Of Classifier Genes From Multi-Class Earthworm Microarray Dataset, Ying Li, Nan Wang, Chaoyang Zhang, Ping Gong

Faculty Publications

Monitoring, assessment and prediction of environmental risks that chemicals pose demand rapid and accurate diagnostic assays. A variety of toxicological effects have been associated with explosive compounds TNT and RDX. One important goal of microarray experiments is to discover novel biomarkers for toxicity evaluation. We have developed an earthworm microarray containing 15,208 unique oligo probes and have used it to profile gene expression in 248 earthworms exposed to TNT, RDX or neither. We assembled a new machine learning pipeline consisting of several well-established feature filtering/selection and classification techniques to analyze the 248-array dataset in order to construct classifier models that …

Event-Driven Similarity And Classification Of Scanpaths, Thomas Grindinger

All Dissertations

Eye tracking experiments often involve recording the pattern of deployment of visual attention over the stimulus as viewers perform a given task (e.g., visual search). It is useful in training applications, for example, to make available an expert's sequence of eye movements, or scanpath, to novices for their inspection and subsequent learning. It may also be potentially useful to be able to assess the conformance of the novice's scanpath to that of the expert. A computational tool is proposed that provides a framework for performing such classification, based on the use of a probabilistic machine learning algorithm. The approach was …

Malware Type Recognition And Cyber Situational Awareness, Thomas E. Dube, Richard A. Raines, Gilbert L. Peterson, Kenneth W. Bauer, Michael R. Grimaila, Steven K. Rogers

Faculty Publications

Current technologies for computer network and host defense do not provide suitable information to support strategic and tactical decision making processes. Although pattern-based malware detection is an active research area, the additional context of the type of malware can improve cyber situational awareness. This additional context is an indicator of threat capability thus allowing organizations to assess information losses and focus response actions appropriately. Malware Type Recognition (MaTR) is a research initiative extending detection technologies to provide the additional context of malware types using only static heuristics. Test results with MaTR demonstrate over a 99% accurate detection rate and 59% …

Analysis Of The Economic Sustainability Of Companies In The Water Sector, Lyubomir Halachev, A. Guergachi, Yashodhan Athavale, Sri Krishnan

International Congress on Environmental Modelling and Software

Many recent studies look at the issue of sustainability from the environmental and the social points of view. This paper takes a different approach and proposes to look at the economic stability of organizations in the water sector. It analyzes a set of 140 companies from around the world involved in this sector, and attempts to develop a methodology to find out, based on stock market data only, which ones will likely remain active in the foreseeable future, and which ones will likely "die" out. This methodology uses machine learning techniques to handle the stock price time series of companies …

Analysis Of The Economic Sustainability Of Companies In The Water Sector, Lyubomir Halachev, A. Guergachi, Yashodhan Athavale, Sri Krishnan

International Congress on Environmental Modelling and Software

Many recent studies look at the issue of sustainability from the environmental and the social points of view. This paper takes a different approach and proposes to look at the economic stability of organizations in the water sector. It analyzes a set of 140 companies from around the world involved in this sector, and attempts to develop a methodology to find out, based on stock market data only, which ones will likely remain active in the foreseeable future, and which ones will likely "die" out. This methodology uses machine learning techniques to handle the stock price time series of companies …

Practical Improvements In Applied Spectral Learning, Adam C. Drake

Theses and Dissertations

Spectral learning algorithms, which learn an unknown function by learning a spectral representation of the function, have been widely used in computational learning theory to prove many interesting learnability results. These algorithms have also been successfully used in real-world applications. However, previous work has left open many questions about how to best use these methods in real-world learning scenarios. This dissertation presents several significant advances in real-world spectral learning. It presents new algorithms for finding large spectral coefficients (a key sub-problem in spectral learning) that allow spectral learning methods to be applied to much larger problems and to a wider …

Transformation Learning: Modeling Transferable Transformations In High-Dimensional Data, Christopher R. Wilson

Theses and Dissertations

The goal of learning transfer is to apply knowledge gained from one problem to a separate related problem. Transformation learning is a proposed approach to computational learning transfer that focuses on modeling high-level transformations that are well suited for transfer. By using a high-level representation of transferable data, transformation learning facilitates both shallow transfer (intra-domain) and deep transfer (inter-domain) scenarios. Transformations can be discovered in data using manifold learning to order data instances according to the transformations they represent. For high-dimensional data representable with coordinate systems, such as images and sounds, data instances can be decomposed into small sub-instances based …

A Comparative Study On Text Categorization, Aditya Chainulu Karamcheti

UNLV Theses, Dissertations, Professional Papers, and Capstones

Automated text categorization is a supervised learning task, defined as assigning category labels to new documents based on likelihood suggested by a training set of labeled documents. Two examples of methodology for text categorizations are Naive Bayes and K-Nearest Neighbor.

In this thesis, we implement two categorization engines based on Naive Bayes and K-Nearest Neighbor methodology. We then compare the effectiveness of these two engines by calculating standard precision and recall for a collection of documents. We will further report on time efficiency of these two engines.

Segmentation And Fracture Detection In X-Ray Images For Traumatic Pelvic Injury, Rebecca Smith

Theses and Dissertations

Due to the risk of complications such as hemorrhage, severe pelvic trauma is associated with a high mortality rate. Prompt medical treatment is therefore vital. However, the complexity of the injuries can make successful diagnosis and treatment challenging. By generating predictions and recommendations based on patient data, computer-aided decision support systems have the potential to assist physicians in improving outcomes. However, no current system considers features automatically extracted from medical images. This dissertation describes a system to extract diagnostic features from pelvic X-ray images that can be used as input to the prediction process; specifically, the presence of fracture and …

Developing Cyberspace Data Understanding Using Crisp-Dm For Host-Based Ids Feature Mining, Joseph R. Erskine, Gilbert L. Peterson, Barry E. Mullins, Michael R. Grimaila

Faculty Publications

Current intrusion detection systems (IDS) generate a large number of specific alerts, but typically do not provide actionable information. Compounding this problem is the fact that many alerts are false positive alerts. This paper applies the Cross Industry Standard Process for Data Mining (CRISP-DM) to develop an understanding of a host environment under attack. Data is generated by launching scans and exploits at a machine outfitted with a set of host-based forensic data collectors. Through knowledge discovery, features are selected to project human understanding of the attack process into the IDS model. By discovering relationships between the data collected and …

Inference Of Cancer-Specific Gene Regulatory Networks Using Soft Computing Rules., Xiaosheng Wang, Osamu Gotoh

Journal Articles: Genetics, Cell Biology & Anatomy

Perturbations of gene regulatory networks are essentially responsible for oncogenesis. Therefore, inferring the gene regulatory networks is a key step to overcoming cancer. In this work, we propose a method for inferring directed gene regulatory networks based on soft computing rules, which can identify important cause-effect regulatory relations of gene expression. First, we identify important genes associated with a specific cancer (colon cancer) using a supervised learning approach. Next, we reconstruct the gene regulatory networks by inferring the regulatory relations among the identified genes, and their regulated relations by other genes within the genome. We obtain two meaningful findings. One …

Extensions Of Nearest Shrunken Centroid Method For Classification, Tomohiko Funai

Theses and Dissertations

Stylometry assumes that the essence of the individual style of an author can be captured using a number of quantitative criteria, such as the relative frequencies of noncontextual words (e.g., or, the, and, etc.). Several statistical methodologies have been developed for authorship analysis. Jockers et al. (2009) utilize Nearest Shrunken Centroid (NSC) classification, a promising classification methodology in DNA microarray analysis for authorship analysis of the Book of Mormon. Schaalje et al. (2010) develop an extended NSC classification to remedy the problem of a missing author. Dabney (2005) and Koppel et al. (2009) suggest other modifications of NSC. This paper …