Computer Sciences Commons^™

A Confidence-Prioritization Approach To Data Processing In Noisy Data Sets And Resulting Estimation Models For Predicting Streamflow Diel Signals In The Pacific Northwest, Nathaniel Lee Gustafson

Theses and Dissertations

Streams in small watersheds are often known to exhibit diel fluctuations, in which streamflow oscillates on a 24-hour cycle. Streamflow diel fluctuations, which we investigate in this study, are an informative indicator of environmental processes. However, in Environmental Data sets, as well as many others, there is a range of noise associated with individual data points. Some points are extracted under relatively clear and defined conditions, while others may include a range of known or unknown confounding factors, which may decrease those points' validity. These points may or may not remain useful for training, depending on how much uncertainty they …

On The K-Mer Frequency Spectra Of Organism Genome And Proteome Sequences With A Preliminary Machine Learning Assessment Of Prime Predictability, Nathan O. Schmidt

Boise State University Theses and Dissertations

A regular expression and region-specific filtering system for biological records at the National Center for Biotechnology database is integrated into an object oriented sequence counting application, and a statistical software suite is designed and deployed to interpret the resulting k-mer frequencies|with a priority focus on nullomers. The proteome k-mer frequency spectra of ten model organisms and the genome k-mer frequency spectra of two bacteria and virus strains for the coding and non-coding regions are comparatively scrutinized. We observe that the naturally-evolved (NCBI/organism) and the artificially-biased (randomly-generated) sequences exhibit a clear deviation from the artificially-unbiased (randomly-generated) histogram distributions. …

On The Automatic Recognition Of Human Activities Using Heterogeneous Wearable Sensors, Oscar David Lara Yejas

USF Tampa Graduate Theses and Dissertations

Delivering accurate and opportune information on people's activities and behaviors has become one of the most important tasks within pervasive computing. Its wide spectrum of potential applications in medical, entertainment, and tactical scenarios, motivates further

research and development of new strategies to improve accuracy, pervasiveness, and eciency.

This dissertation addresses the recognition of human activities (HAR) with wearable sensors in three main regards: In the rst place, physiological signals have been incorporated as a new source of information to improve the recognition accuracy achieved by conventional approaches, which rely on accelerometer signals solely. A new HAR system, Centinela, was born …

Ensemble Methods For Malware Diagnosis Based On One-Class Svms, Xing An

LSU Master's Theses

Malware diagnosis is one of today’s most popular topics of machine learning. Instead of simply applying all the classical classification algorithms to the problem and claim the highest accuracy as the result of prediction, which is the typical approach adopted by studies of this kind, we stick to the Support Vector Machine (SVM) classifier and based on our observation of some principles of learning, characteristics of statistics and the behavior of SVM, we employed a number of the potential preprocessing or ensemble methods including rescaling, bagging and clustering that may enhance the performance to the classical algorithm. We implemented the …

A Study Of Localization And Latency Reduction For Action Recognition, Syed Zain Masood

Electronic Theses and Dissertations

The success of recognizing periodic actions in single-person-simple-background datasets, such as Weizmann and KTH, has created a need for more complex datasets to push the performance of action recognition systems. In this work, we create a new synthetic action dataset and use it to highlight weaknesses in current recognition systems. Experiments show that introducing background complexity to action video sequences causes a significant degradation in recognition performance. Moreover, this degradation cannot be fixed by fine-tuning system parameters or by selecting better feature points. Instead, we show that the problem lies in the spatio-temporal cuboid volume extracted from the interest point …

Software Process Evaluation: A Machine Learning Approach, Ning Chen, Steven C. H. Hoi, Xiaokui Xiao

Research Collection School Of Computing and Information Systems

Software process evaluation is essential to improve software development and the quality of software products in an organization. Conventional approaches based on manual qualitative evaluations (e.g., artifacts inspection) are deficient in the sense that (i) they are time-consuming, (ii) they suffer from the authority constraints, and (iii) they are often subjective. To overcome these limitations, this paper presents a novel semi-automated approach to software process evaluation using machine learning techniques. In particular, we formulate the problem as a sequence classification task, which is solved by applying machine learning algorithms. Based on the framework, we define a new quantitative indicator to …

Active Multiple Kernel Learning For Interactive 3d Object Retrieval Systems, Steven C. H. Hoi, Rong Jin

Research Collection School Of Computing and Information Systems

An effective relevance feedback solution plays a key role in interactive intelligent 3D object retrieval systems. In this work, we investigate the relevance feedback problem for interactive intelligent 3D object retrieval, with the focus on studying effective machine learning algorithms for improving the user's interaction in the retrieval task. One of the key challenges is to learn appropriate kernel similarity measure between 3D objects through the relevance feedback interaction with users. We address this challenge by presenting a novel framework of Active multiple kernel learning (AMKL), which exploits multiple kernel learning techniques for relevance feedback in interactive 3D object retrieval. …

Active Multiple Kernel Learning For Interactive 3d Object Retrieval Systems, Steven C. H. Hoi, Rong Jin

Research Collection School Of Computing and Information Systems

An effective relevance feedback solution plays a key role in interactive intelligent 3D object retrieval systems. In this work, we investigate the relevance feedback problem for interactive intelligent 3D object retrieval, with the focus on studying effective machine learning algorithms for improving the user's interaction in the retrieval task. One of the key challenges is to learn appropriate kernel similarity measure between 3D objects through the relevance feedback interaction with users. We address this challenge by presenting a novel framework of Active multiple kernel learning (AMKL), which exploits multiple kernel learning techniques for relevance feedback in interactive 3D object retrieval. …

Implementation Of A New Sigmoid Function In Backpropagation Neural Networks., Jeffrey A. Bonnell

Electronic Theses and Dissertations

This thesis presents the use of a new sigmoid activation function in backpropagation artificial neural networks (ANNs). ANNs using conventional activation functions may generalize poorly when trained on a set which includes quirky, mislabeled, unbalanced, or otherwise complicated data. This new activation function is an attempt to improve generalization and reduce overtraining on mislabeled or irrelevant data by restricting training when inputs to the hidden neurons are sufficiently small. This activation function includes a flattened, low-training region which grows or shrinks during back-propagation to ensure a desired proportion of inputs inside the low-training region. With a desired low-training proportion of …

Development Of Advanced Algorithms To Detect, Characterize And Forecast Solar Activities, Yuan Yuan

Dissertations

Study of the solar activity is an important part of space weather research. It is facing serious challenges because of large data volume, which requires application of state-of-the-art machine learning and computer vision techniques. This dissertation targets at two essential aspects in space weather research: automatic feature detection and forecasting of eruptive events.

Feature detection includes solar filament detection and solar fibril tracing. A solar filament consists of a mass of gas suspended over the chromosphere by magnetic fields and seen as a dark, ribbon-shaped feature on the bright solar disk in Hα (Hydrogen-alpha) full-disk solar images. In this dissertation, …

Hardware Acceleration Of Inference Computing: The Numenta Htm Algorithm, Dan Hammerstrom

Systems Science Friday Noon Seminar Series

In this presentation I will describe the latest version of the Numenta HTM Cortical Learning Algorithm and why it is interesting for doing research into radical new computer architectures. Then I will discuss the hardware acceleration research we are doing, and briefly look at some preliminary applications development.

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 …

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 …

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 …

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 …

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 …

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 …

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% …

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 …

A Bayesian Decision Theoretical Approach To Supervised Learning, Selective Sampling, And Empirical Function Optimization, James Lamond Carroll

Theses and Dissertations

Many have used the principles of statistics and Bayesian decision theory to model specific learning problems. It is less common to see models of the processes of learning in general. One exception is the model of the supervised learning process known as the "Extended Bayesian Formalism" or EBF. This model is descriptive, in that it can describe and compare learning algorithms. Thus the EBF is capable of modeling both effective and ineffective learning algorithms. We extend the EBF to model un-supervised learning, semi-supervised learning, supervised learning, and empirical function optimization. We also generalize the utility model of the EBF to …