Computer Sciences Commons^™

Comparison Of Different Methods For Determining Diabetes, Mehmet Recep Bozkurt, Ni̇lüfer Yurtay, Zi̇ynet Yilmaz, Cengi̇z Sertkaya

Turkish Journal of Electrical Engineering and Computer Sciences

In this study, the Pima Indian Diabetes dataset was categorized with 8 different classifiers. The data were taken from the University of California Irvine Machine Learning Repository's web site. As a classifier, 6 different neural networks [probabilistic neural network (PNN), learning vector quantization, feedforward networks, cascade-forward networks, distributed time delay networks (DTDN), and time delay networks], the artificial immune system, and the Gini algorithm from decision trees were used. The classifier's performance ratios were studied separately as accuracy, sensitivity, and specificity and the success rates of all of the classifiers are presented. Among these 8 classifiers, the best accuracy and …

Online Feature Selection And Classification With Incomplete Data, Habi̇l Kalkan

Turkish Journal of Electrical Engineering and Computer Sciences

This paper presents a classification system in which learning, feature selection, and classification for incomplete data are simultaneously carried out in an online manner. Learning is conducted on a predefined model including the class-dependent mean vectors and correlation coefficients, which are obtained by incrementally processing the incoming observations with missing features. A nearest neighbor with a Gaussian mixture model, whose parameters are also estimated from the trained model, is used for classification. When a testing observation is received, the algorithm discards the missing attributes on the observation and ranks the available features by performing feature selection on the model that …

Determining What Characteristics Constitute A Darknet, Symon Aked, Christopher Bolan, Murray Brand

Australian Information Security Management Conference

Privacy on the Internet has always been a concern, but monitoring of content by both private corporations and Government departments has pushed people to search for ways to communicate over the Internet in a more secure manner. This has given rise to the creations of Darknets, which are networks that operate “inside” the Internet, and allow anonymous participation via a de‐centralised, encrypted, peer‐to‐peer network topology. This research investigates some sources of known Internet content monitoring, and how they provided the template for the creation of a system to avoid such surveillance. It then highlights how communications on the Clearnet is …

Predictive Handling Of Asynchronous Concept Drifts In Distributed Environments, Hock Hee Ang, Vivek Gopalkrishnan, Indre Zliobaite, Mykola Pechenizkiy, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

In a distributed computing environment, peers collaboratively learn to classify concepts of interest from each other. When external changes happen and their concepts drift, the peers should adapt to avoid increase in misclassification errors. The problem of adaptation becomes more difficult when the changes are asynchronous, i.e., when peers experience drifts at different times. We address this problem by developing an ensemble approach, PINE, that combines reactive adaptation via drift detection, and proactive handling of upcoming changes via early warning and adaptation across the peers. With empirical study on simulated and real-world data sets, we show that PINE handles asynchronous …

Will Fault Localization Work For These Failures? An Automated Approach To Predict Effectiveness Of Fault Localization Tools, Tien-Duy B. Le, David Lo

Research Collection School Of Computing and Information Systems

Debugging is a crucial yet expensive activity to improve the reliability of software systems. To reduce debugging cost, various fault localization tools have been proposed. A spectrum-based fault localization tool often outputs an ordered list of program elements sorted based on their likelihood to be the root cause of a set of failures (i.e., their suspiciousness scores). Despite the many studies on fault localization, unfortunately, however, for many bugs, the root causes are often low in the ordered list. This potentially causes developers to distrust fault localization tools. Recently, Parnin and Orso highlight in their user study that many debuggers …

An Investigation Of Decision Analytic Methodologies For Stress Identification, Yong Deng, Chao-Hsien Chu, Huayou Si, Qixun Zhang, Zhonghai Wu

Research Collection School Of Computing and Information Systems

In modern society, more and more people are suffering from some type of stress. Monitoring and timely detecting of stress level will be very valuable for the person to take counter measures. In this paper, we investigate the use of decision analytics methodologies to detect stress. We present a new feature selection method based on the principal component analysis (PCA), compare three feature selection methods, and evaluate five information fusion methods for stress detection. A driving stress data set created by the MIT Media lab is used to evaluate the relative performance of these methods. Our study show that the …

Assessment And Prediction Of Cardiovascular Status During Cardiac Arrest Through Machine Learning And Dynamical Time-Series Analysis, Sharad Shandilya

Theses and Dissertations

In this work, new methods of feature extraction, feature selection, stochastic data characterization/modeling, variance reduction and measures for parametric discrimination are proposed. These methods have implications for data mining, machine learning, and information theory. A novel decision-support system is developed in order to guide intervention during cardiac arrest. The models are built upon knowledge extracted with signal-processing, non-linear dynamic and machine-learning methods. The proposed ECG characterization, combined with information extracted from PetCO2 signals, shows viability for decision-support in clinical settings. The approach, which focuses on integration of multiple features through machine learning techniques, suits well to inclusion of multiple physiologic …

Mkboost: A Framework Of Multiple Kernel Boosting, Hao Xia, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Multiple kernel learning (MKL) is a promising family of machine learning algorithms using multiple kernel functions for various challenging data mining tasks. Conventional MKL methods often formulate the problem as an optimization task of learning the optimal combinations of both kernels and classifiers, which usually results in some forms of challenging optimization tasks that are often difficult to be solved. Different from the existing MKL methods, in this paper, we investigate a boosting framework of MKL for classification tasks, i.e., we adopt boosting to solve a variant of MKL problem, which avoids solving the complicated optimization tasks. Specifically, we present …

Geometric Approach To Support Vector Machines Learning For Large Datasets, Robert Strack

Theses and Dissertations

The dissertation introduces Sphere Support Vector Machines (SphereSVM) and Minimal Norm Support Vector Machines (MNSVM) as the new fast classification algorithms that use geometrical properties of the underlying classification problems to efficiently obtain models describing training data. SphereSVM is based on combining minimal enclosing ball approach, state of the art nearest point problem solvers and probabilistic techniques. The blending of the three speeds up the training phase of SVMs significantly and reaches similar (i.e., practically the same) accuracy as the other classification models over several big and large real data sets within the strict validation frame of a double (nested) …

Online Multiple Kernel Classification, Steven C. H. Hoi, Rong Jin, Peilin Zhao, Tianbao Yang

Research Collection School Of Computing and Information Systems

Although both online learning and kernel learning have been studied extensively in machine learning, there is limited effort in addressing the intersecting research problems of these two important topics. As an attempt to fill the gap, we address a new research problem, termed Online Multiple Kernel Classification (OMKC), which learns a kernel-based prediction function by selecting a subset of predefined kernel functions in an online learning fashion. OMKC is in general more challenging than typical online learning because both the kernel classifiers and the subset of selected kernels are unknown, and more importantly the solutions to the kernel classifiers and …

Context Aware Privacy Preserving Clustering And Classification, Nirmal Thapa

Theses and Dissertations--Computer Science

Data are valuable assets to any organizations or individuals. Data are sources of useful information which is a big part of decision making. All sectors have potential to benefit from having information. Commerce, health, and research are some of the fields that have benefited from data. On the other hand, the availability of the data makes it easy for anyone to exploit the data, which in many cases are private confidential data. It is necessary to preserve the confidentiality of the data. We study two categories of privacy: Data Value Hiding and Data Pattern Hiding. Privacy is a huge concern …

Knowledge Extraction From Survey Data Using Neural Networks, Khan Imran, Arun Kulkarni

Computer Science Faculty Publications and Presentations

Surveys are an important tool for researchers. It is increasingly important to develop powerful means for analyzing such data and to extract knowledge that could help in decision-making. Survey attributes are typically discrete data measured on a Likert scale. The process of classification becomes complex if the number of survey attributes is large. Another major issue in Likert-Scale data is the uniqueness of tuples. A large number of unique tuples may result in a large number of patterns. The main focus of this paper is to propose an efficient knowledge extraction method that can extract knowledge in terms of rules. …

A Convex Optimization Algorithm For Sparse Representation And Applications In Classification Problems, Reinaldo Sanchez Arias

Open Access Theses & Dissertations

In pattern recognition and machine learning, a classification problem refers to finding an algorithm for assigning a given input data into one of several categories. Many natural signals are sparse or compressible in the sense that they have short representations when expressed in a suitable basis. Motivated by the recent successful development of algorithms for sparse signal recovery, we apply the selective nature of sparse representation to perform classification. Any test sample is represented in an overcomplete dictionary with the training sample as base elements. A given test sample can be expressed as a linear combination of only those training …

A Rule Induction Algorithm For Knowledge Discovery And Classification, Ömer Akgöbek

Turkish Journal of Electrical Engineering and Computer Sciences

Classification and rule induction are key topics in the fields of decision making and knowledge discovery. The objective of this study is to present a new algorithm developed for automatic knowledge acquisition in data mining. The proposed algorithm has been named RES-2 (Rule Extraction System). It aims at eliminating the pitfalls and disadvantages of the techniques and algorithms currently in use. The proposed algorithm makes use of the direct rule extraction approach, rather than the decision tree. For this purpose, it uses a set of examples to induce general rules. In this study, 15 datasets consisting of multiclass values with …

Hybrid Of Genetic Algorithm And Great Deluge Algorithm For Rough Set Attribute Reduction, Najmeh Sadat Jaddi, Salwani Abdullah

Turkish Journal of Electrical Engineering and Computer Sciences

The attribute reduction problem is the process of reducing unimportant attributes from a decision system to decrease the difficulty of data mining or knowledge discovery tasks. Many algorithms have been used to optimize this problem in rough set theory. The genetic algorithm (GA) is one of the algorithms that has already been applied to optimize this problem. This paper proposes 2 kinds of memetic algorithms, which are a hybridization of the GA, with 2 versions (linear and nonlinear) of the great deluge (GD) algorithm. The purpose of this hybridization is to investigate the ability of this local search algorithm to …

Detection Of Microcalcification Clusters In Digitized X-Ray Mammograms Using Unsharp Masking And Image Statistics, Peli̇n Kuş, İrfan Karagöz

Turkish Journal of Electrical Engineering and Computer Sciences

A fully automated method for detecting microcalcification (MC) clusters in regions of interest (ROIs) extracted from digitized X-ray mammograms is proposed. In the first stage, an unsharp masking is used to perform the contrast enhancement of the MCs. In the second stage, the ROIs are decomposed into a 2-level contourlet representation and the reconstruction is obtained by eliminating the low-frequency subband in the second level. In the third stage, statistical textural features are extracted from the ROIs and they are classified using support vector machines. To test the performance of the method, 57 ROIs selected from the Mammographic Image Analysis …

Exploitation Of Rf-Dna For Device Classification And Verification Using Grlvqi Processing, Donald R. Reising

Theses and Dissertations

This dissertation introduces a GRLVQI classifier into an RF-DNA fingerprinting process and demonstrates applicability for device classification and ID verification. Unlike MDA/ML processing, GRLVQI provides a measure of feature relevance that enables Dimensional Reduction Analysis (DRA) to enhance the experimental-to-operational transition potential of RF-DNA fingerprinting. Using 2D Gabor Transform RF-DNA fingerprints extracted from experimentally collected OFDM-based 802.16 WiMAX and 802.11 WiFi device emissions, average GRLVQI classification accuracy of %C greater than or equal to 90% is achieved using full and reduced dimensional feature sets at SNR greater than or equal to 10.0 dB and SNR greater than or equal to …

Cost-Sensitive Online Classification, Jialei Wang, Peilin Zhao, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Both cost-sensitive classification and online learning have been extensively studied in data mining and machine learning communities, respectively. However, very limited study addresses an important intersecting problem, that is, “Cost-Sensitive Online Classification". In this paper, we formally study this problem, and propose a new framework for Cost-Sensitive Online Classification by directly optimizing cost-sensitive measures using online gradient descent techniques. Specifically, we propose two novel cost-sensitive online classification algorithms, which are designed to directly optimize two well-known cost-sensitive measures: (i) maximization of weighted sum of sensitivity and specificity, and (ii) minimization of weighted misclassification cost. We analyze the theoretical bounds of …

Fast And Efficient Classification, Tracking, And Simulation In Wireless Sensor Networks, Hao Jiang

All Dissertations

Wireless sensor networks are composed of large numbers of resource-lean sensors that collect low-level inputs from the physical world. The applications present challenges for programmers. On the one hand, lightweight algorithms are required given the limited capacity of the constituent devices. On the other, the algorithms must be scalable to accommodate large networks. In this thesis, we focus on the design and implementation of fast and lean (yet scalable) algorithms for classification, simulation, and target tracking in the context of wireless sensor networks. We briefly consider each of these challenges in turn.
The first challenge is to achieve high precision …

Online Feature Selection For Mining Big Data, Steven C. H. Hoi, Jialei Wang, Peilin Zhao, Rong Jin

Research Collection School Of Computing and Information Systems

Most studies of online learning require accessing all the attributes/features of training instances. Such a classical setting is not always appropriate for real-world applications when data instances are of high dimensionality or the access to it is expensive to acquire the full set of attributes/features. To address this limitation, we investigate the problem of Online Feature Selection (OFS) in which the online learner is only allowed to maintain a classifier involved a small and fixed number of features. The key challenge of Online Feature Selection is how to make accurate prediction using a small and fixed number of active features. …

Recognizing Patterns In Transmitted Signals For Identification Purposes, Baha' A. Alsaify

Graduate Theses and Dissertations

The ability to identify and authenticate entities in cyberspace such as users, computers, cell phones, smart cards, and radio frequency identification (RFID) tags is usually accomplished by having the entity demonstrate knowledge of a secret key. When the entity is portable and physically accessible, like an RFID tag, it can be difficult to secure given the memory, processing, and economic constraints. This work proposes to use unique patterns in the transmitted signals caused by manufacturing differences to identify and authenticate a wireless device such as an RFID tag. Both manufacturer identification and tag identification are performed on a population of …

Composite Feature-Based Face Detection Using Skin Color Modeling And Svm Classification, Swathi Rajashekar

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

This report proposes a face detection algorithm based on skin color modeling and support vector machine (SVM) classification. Said classification is based on various face features used to detect specific faces in an input color image. A YCbCr color space is used to filter the skin color pixels from the input color image. Template matching is used on the result with various window sizes of the template created from an ORL face database. The candidates obtained above, are then classified by SVM classifiers using the histogram of oriented gradients, eigen features, edge ratio, and edge statistics features.

Fast Neural Network Algorithm For Solving Classification Tasks, Noor Albarakati

Theses and Dissertations

Classification is one-out-of several applications in the neural network (NN) world. Multilayer perceptron (MLP) is the common neural network architecture which is used for classification tasks. It is famous for its error back propagation (EBP) algorithm, which opened the new way for solving classification problems given a set of empirical data. In the thesis, we performed experiments by using three different NN structures in order to find the best MLP neural network structure for performing the nonlinear classification of multiclass data sets. A developed learning algorithm used here is the batch EBP algorithm which uses all the data as a …

Contributions To K-Means Clustering And Regression Via Classification Algorithms, Raied Salman

Theses and Dissertations

The dissertation deals with clustering algorithms and transforming regression prob-lems into classification problems. The main contributions of the dissertation are twofold; first, to improve (speed up) the clustering algorithms and second, to develop a strict learn-ing environment for solving regression problems as classification tasks by using support vector machines (SVMs). An extension to the most popular unsupervised clustering meth-od, k-means algorithm, is proposed, dubbed k-means2 (k-means squared) algorithm, appli-cable to ultra large datasets. The main idea is based on using a small portion of the dataset in the first stage of the clustering. Thus, the centers of such a smaller …

Computer Methods For Pre-Microrna Secondary Structure Prediction, Dianwei Han

Theses and Dissertations--Computer Science

This thesis presents a new algorithm to predict the pre-microRNA secondary structure. An accurate prediction of the pre-microRNA secondary structure is important in miRNA informatics. Based on a recently proposed model, nucleotide cyclic motifs (NCM), to predict RNA secondary structure, we propose and implement a Modified NCM (MNCM) model with a physics-based scoring strategy to tackle the problem of pre-microRNA folding. Our microRNAfold is implemented using a global optimal algorithm based on the bottom-up local optimal solutions.

It has been shown that studying the functions of multiple genes and predicting the secondary structure of multiple related microRNA is more important …

Assessing Metacognitive Skills Using Adaptive Neural Networks, Anderson Justin, Kouider Mokhtari, Arun Kulkarni

Computer Science Faculty Publications and Presentations

The assessment of student's levels of metacognitive knowledge and skills is critical in determining their ability to effectively perform complex cognitive tasks such as solving mathematics or reading comprehension problems. In this paper, we use an adaptive multiplayer perceptron model to categorize participants based on their metacognitive awareness and perceived use of reading strategies while reading. Eight hundred and sixty-five middle school students participated in the study. All participants completed a 30-item instrument- the Metacognitive Awareness-of-Reading Strategies Inventory (MARSI). We used adaptive multi-layer perceptron models to classify participants into three groups based on their metacognitive strategy awareness levels using thirteen …

Demographic Prediction Of Mobile User From Phone Usage, Shahram Mohrehkesh, Shuiwang Ji, Tamer Nadeem, Michele C. Weigle

Computer Science Faculty Publications

In this paper, we describe how we use the mobile phone usage of users to predict their demographic attributes. Using call log, visited GSM cells information, visited Bluetooth devices, visited Wireless LAN devices, accelerometer data, and so on, we predict the gender, age, marital status, job and number of people in household of users. The accuracy of developed classifiers for these classification problems ranges from 45-87% depending upon the particular classification problem.

Stability And Classification Performance Of Feature Selection Techniques, Huanjing Wang, Taghi Khoshgoftaar, Qianhui Liang

Computer Science Faculty Publications

Feature selection techniques can be evaluated based on either model performance or the stability (robustness) of the technique. The ideal situation is to choose a feature selec- tion technique that is robust to change, while also ensuring that models built with the selected features perform well. One domain where feature selection is especially important is software defect prediction, where large numbers of met- rics collected from previous software projects are used to help engineers focus their efforts on the most faulty mod- ules. This study presents a comprehensive empirical ex- amination of seven filter-based feature ranking techniques (rankers) applied to …

Collaborative Online Learning Of User Generated Content, Guangxia Li, Kuiyu Chang, Steven C. H. Hoi, Wenting Liu, Ramesh Jain

Research Collection School Of Computing and Information Systems

We study the problem of online classification of user generated content, with the goal of efficiently learning to categorize content generated by individual user. This problem is challenging due to several reasons. First, the huge amount of user generated content demands a highly efficient and scalable classification solution. Second, the categories are typically highly imbalanced, i.e., the number of samples from a particular useful class could be far and few between compared to some others (majority class). In some applications like spam detection, identification of the minority class often has significantly greater value than that of the majority class. Last …

Using Semantic Templates To Study Vulnerabilities Recorded In Large Software Repositories, Yan Wu

Student Work

Software vulnerabilities allow an attacker to reduce a system's Confidentiality, Availability, and Integrity by exposing information, executing malicious code, and undermine system functionalities that contribute to the overall system purpose and need. With new vulnerabilities discovered everyday in a variety of applications and user environments, a systematic study of their characteristics is a subject of immediate need for the following reasons:

• The high rate in which information about past and new vulnerabilities are accumulated makes it difficult to absorb and comprehend.
• Rather than learning from past mistakes, similar types of vulnerabilities are observed repeatedly.
• As the scale and complexity of …