Computer Sciences Commons^™

Introducing Machine Learning Via Baseball's Hall Of Fame, David Hansen

Faculty Publications - Department of Electrical Engineering and Computer Science

Machine Learning via Artificial Neural Networks (ANNs) is often introduced in a one-semester course on Artificial Intelligence. Baseball’s annual Hall of Fame election provides a simple, tractable, data-rich domain for learning how to use ANNs for predictive analytics. We describe how we use the Fast Artificial Neural Network (FANN) toolkit for a course assignment that predicts which players are likely to be elected to Baseball’s Hall of Fame.

Access Control Programming Library And Exploration System, Zhitao Qiu

Dissertations, Master's Theses and Master's Reports - Open

The high complexity of advanced security models in the modern trusted systems requires an effective formal education for students. Education access control tools have been promoted. Though they can benefit the learning through analyzing or visualizing access control policies, few of them are designed to teach development of access control policies.

In this report, we propose an access control programming library which can provide students hand-on experience with the effect of an access control policy on a running program. A student can write a policy and then run programs under the policy. The Programming Library provides a system call wrapper …

Serve Or Skip: The Power Of Rejection In Online Bottleneck Matching, Barbara M. Anthony, Christine Chung

Computer Science Faculty Publications

We consider the online matching problem, where n server-vertices lie in a metric space and n request-vertices that arrive over time each must immediately be permanently assigned to a server-vertex.We focus on the egalitarian bottleneck objective, where the goal is to minimize the maximum distance between any request and its server. It has been demonstrated that while there are effective algorithms for the utilitarian objective (minimizing total cost) in the resource augmentation setting where the offline adversary has half the resources, these are not effective for the egalitarian objective. Thus, we propose a new Serve-or-Skip bicriteria analysis model, where the …

Numerical Investigation On Charring Ablator Geometric Effects: Study Of Stardust Sample Return Capsule Heat Shield, Haoyue Weng, Alexandre Martin

Mechanical Engineering Faculty Publications

Sample geometry is very influential in small charring ablative articles where 1D assumption might not be accurate. In heat shield design, 1D is often assumed since the nose radius is much larger than the thickness of charring. Whether the 1D assumption is valid for the heat shield is unknown. Therefore, the geometric effects of Stardust sample return capsule heat shield are numerically studied using a material response program. The developed computer program models material charring, conductive heat transfer, surface energy balance, pyrolysis gas transport and orthotropic material properties in 3D Cartesian coordinates. Simulation results show that the centerline temperatures predicted …

Using Regression Tools To Assess Hazard Identification In The U.S. Army Risk Management Process, Rania Hodhod, Heath L. Mccormick

Faculty Bibliography

This research considers whether a person‟s demographic and experiential attributes play a significant role in how they perceive the presence or absence of hazards in a given situation. The goal of the research is to show that participants with enlisted military experience, prior to being commissioned as a junior officer, would be more successful at identifying the hazards presented in military scenarios than those who had only been trained on the process via their pre-commissioning and initial entry courses of instruction. The research study involves the use of two surveys with realistic military scenarios including both Foot March and Maintenance …

An Evaluation-Guided Approach For Effective Data Visualization On Tablets, Peter S. Games, Alark Joshi

Computer Science

There is a rising trend of data analysis and visualization tasks being performed on a tablet device. Apps with interactive data visualization capabilities are available for a wide variety of domains. We investigate whether users grasp how to effectively interpret and interact with visualizations. We conducted a detailed user evaluation to study the abilities of individuals with respect to analyzing data on a tablet through an interactive visualization app. Based upon the results of the user evaluation, we find that most subjects performed well at understanding and interacting with simple visualizations, specifically tables and line charts. A majority of the …

Ip Basics: Copyright For Digital Authors, Thomas G. Field Jr.

Law Faculty Scholarship

Written for computer artists and programmers, this paper addresses the basics, as well as the registration of multiple works, difference between works that are and are not prepared "for hire," and other matters of interest to entrepreneurs as well as to free-lance programmers and artists.

Injective Type Families For Haskell (Extended Version), Jan Stolarek, Simon Peyton Jones, Richard A. Eisenberg

Computer Science Faculty Research and Scholarship

Haskell, as implemented by the Glasgow Haskell Compiler (GHC), allows expressive type-level programming. The most popular type- level programming extension is TypeFamilies, which allows users to write functions on types. Yet, using type functions can cripple type inference in certain situations. In particular, lack of injectivity in type functions means that GHC can never infer an instantiation of a type variable appearing only under type functions.

In this paper, we describe a small modification to GHC that allows type functions to be annotated as injective. GHC naturally must check validity of the injectivity annotations. The algorithm to do so is …

An Overabundance Of Equality: Implementing Kind Equalities Into Haskell, Richard A. Eisenberg

Computer Science Faculty Research and Scholarship

Haskell, as embodied by version 7.10.1 of the Glasgow Haskell Compiler (GHC), supports reasoning about equality among types, via generalized algebraic datatypes (GADTs) and type families. However, these features are not available among the kinds that clas- sify the types. Motivated by a concrete example of how kind equali- ties can help programmers today, this paper presents the challenges and solutions encountered in integrating kind equalities into GHC, an industrial-strength compiler. The challenges addressed here all surround the many notions of type equality that exist in GHC to- day, and in particular around GHC’s role mechanism. These many different relations …

Quick Git Setup, Joseph Lawrence, Seikyung Jung

Computer Science Faculty Publications

Version control is widely adopted in industry because it enables software development in groups, yet few students gain sufficient experience through their undergraduate courses. Even though version control is ideal for work submission, faculty may avoid it in favor of course management systems used only in academia. This tutorial introduces software to automate setting up version control with cloud project hosting services, and gives experience with version control as a side-effect of work submission and collection. This tutorial assumes no prior experience.

Security Assessment Of Iot Devices: The Case Of Two Smart Tvs, Maxim Chernyshev, Peter Hannay

Australian Digital Forensics Conference

Being increasingly complex devices, smart TVs are becoming more capable and have the potential to receive, store, process and transmit considerable amounts of personal data. These capabilities also represent several diverse attack surfaces potentially rendering these devices highly vulnerable. The emergence and high adoption rate of smart TVs have been drawing notable interest from security researchers and industry. We utilise an attack surface area-based approach to assess the security of two modern smart TVs from different vendors and describe some of the possible multi-surface attacks that can be carried out against these devices.

Direct Optimization For Classification With Boosting, Shaodan Zhai

Browse all Theses and Dissertations

Boosting, as one of the state-of-the-art classification approaches, is widely used in the industry for a broad range of problems. The existing boosting methods often formulate classification tasks as a convex optimization problem by using surrogates of performance measures. While the convex surrogates are computationally efficient to globally optimize, they are sensitive to outliers and inconsistent under some conditions. On the other hand, boosting's success can be ascribed to maximizing the margins, but few boosting approaches are designed to directly maximize the margin. In this research, we design novel boosting algorithms that directly optimize non-convex performance measures, including the empirical …

Contrast Pattern Aided Regression And Classification, Vahid Taslimitehrani

Browse all Theses and Dissertations

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

Potential Of Cognitive Computing And Cognitive Systems, Ahmed K. Noor

Computational Modeling & Simulation Engineering Faculty Publications

Cognitive computing and cognitive technologies are game changers for future engineering systems, as well as for engineering practice and training. They are major drivers for knowledge automation work, and the creation of cognitive products with higher levels of intelligence than current smart products. This paper gives a brief review of cognitive computing and some of the cognitive engineering systems activities. The potential of cognitive technologies is outlined, along with a brief description of future cognitive environments, incorporating cognitive assistants - specialized proactive intelligent software agents designed to follow and interact with humans and other cognitive assistants across the environments. The …

Comparison Of Live Response, Linux Memory Extractor (Lime) And Mem Tool For Acquiring Android’S Volatile Memory In The Malware Incident, Andri Heriyanto, Craig Valli, Peter Hannay

Australian Digital Forensics Conference

The increasing use of encryption and obfuscation within the malware development arena has necessitated the use of volatile memory acquisition on smartphone platforms. Current smartphone forensics research lacks a well-formulated process for the acquisition of volatile memory. This research evaluates and contrasts three differing tools for acquisition of volatile memory from the Android platform: Live Response, Linux Memory Extractor (LiME) and Mem Tool. Evaluation is conducted through practical examination during the analysis of an infected device. The results demonstrate a combination of LiME and the Volatility Framework provides the most robust findings. Complexities due to the nature of LiME prevent …

Mining Social Networking Sites For Digital Evidence, Brian Cusack, Saud Alshaifi

Australian Digital Forensics Conference

OnLine Social Networking sites (SNS) hold a vast amount of information that individuals and organisations post about themselves. Investigations include SNS as sources of evidence and the challenge is to have effective tools to extract the evidence. In this exploratory research we apply the latest version of a proprietary tool to identify potential evidence from five SNS using three different browsers. We found that each web browser influenced the scope of the evidence extracted. In previous research we have shown that different open source and proprietary tools influence the scope of evidence obtained. In this research we asked, What variation …

Respiratory Particle Deposition Probability Due To Sedimentation With Variable Gravity And Electrostatic Forces, Ioannis Haranas, Ioannis Gkigkitzis, George D. Zouganelis, Maria K. Haranas, Samantha Kirk

Physics and Computer Science Faculty Publications

In this paper, we study the effects of the acceleration gravity on the sedimentation deposition probability, as well as the aerosol deposition rate on the surface of the Earth and Mars, but also aboard a spacecraft in orbit around Earth and Mars as well for particles with density ρ_p = 1300 kg/m³, diameters d_p = 1, 3, 5 µm and residence times t = 0.0272, 0.2 s respectively. For particles of diameter 1 µm we find that, on the surface of Earth and Mars the deposition probabilities are higher at the poles when compared to the …

Steganography As A Threat – Fairytale Or Fact?, Tom Cleary

Australian Digital Forensics Conference

Almost since the birth of the Internet, there has been a fear that steganographically-encoded threats would be used to bring harm. Serious consideration has been given to the idea that merely downloading an image could introduce malware. Yet, for decades, evidence of this malware channel has been missing in action. There is still an unwritten assumption that images are harmless. Many vendors have implicitly avoided producing defences against steganographic threats. Is it truly impossible to make a widely harmful exploit this way or have malicious actors accepted general wisdom? Three recent papers suggest that there may be a new chapter …

Optimizing Pred(25) Is Np-Hard, Martine Ceberio, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

Usually, in data processing, to find the parameters of the models that best fits the data, people use the Least Squares method. One of the advantages of this method is that for linear models, it leads to an easy-to-solve system of linear equations. A limitation of this method is that even a single outlier can ruin the corresponding estimates; thus, more robust methods are needed. In particular, in software engineering, often, a more robust pred(25) method is used, in which we maximize the number of cases in which the model's prediction is within the 25% range of the observations. In …

Music-Related Media-Contents Synchronization Over Theweb: The Ieee 1599 Initiative, Adriano Baratè, Goffredo Haus, Luca A. Ludovico, Stefano Baldan, Davide Andrea Mauro

Computer Sciences and Electrical Engineering Faculty Research

IEEE 1599 is an international standard originally conceived for music, which aims at providing a comprehensive description of the media contents related to a music piece within a multi-layer and synchronized environment. A number of o_- line and stand-alone software prototypes has been realized after its standardization, occurred in 2008. Recently, thanks to some technological advances (e.g. the release of HTML5), the engine of the IEEE 1599 parser has been ported on the Web. Some non-trivial problems have been solved, e.g. the management of multiple simultaneous media streams in a client-server architecture. After providing an overview of the IEEE 1599 …

Isolation In Synchronized Drone Formations, Andrew P. Brunner

Electronic Theses and Dissertations

This paper expands on a theoretical model that is used for aerial robots that are working cooperatively to complete a task. In certain situations, such as when multiple robots have catastrophic failures, the surviving robots could become isolated so that they never again communicate with another robot. We prove some properties about isolated robots flying in a grid formation, and we present an algorithm that determines how many robots need to fail to isolate at least one robot. Finally, we propose a strategy that eliminates the possibility of isolation altogether.

Facial Expression Analysis Via Transfer Learning, Xiao Zhang

Electronic Theses and Dissertations

Automated analysis of facial expressions has remained an interesting and challenging research topic in the field of computer vision and pattern recognition due to vast applications such as human-machine interface design, social robotics, and developmental psychology. This dissertation focuses on developing and applying transfer learning algorithms - multiple kernel learning (MKL) and multi-task learning (MTL) - to resolve the problems of facial feature fusion and the exploitation of multiple facial action units (AUs) relations in designing robust facial expression recognition systems. MKL algorithms are employed to fuse multiple facial features with different kernel functions and tackle the domain adaption problem …

The Power Of Technology In Cre Data And Analytics, Clarence Goh

Research Collection School of Accountancy

Many companies are using data to drive competitiveadvantage. Across industries, there is rapidly growingappreciation that data-driven insights can substantiallyimprove decision making across a wide range of businessfunctions, and corporate real estate (CRE) is no exception.

Efficient Training Of Small Kernel Convolutional Neural Networks Using Fast Fourier Transform, Tyler Highlander

Browse all Theses and Dissertations

Convolutional neural networks (CNNs) are currently state-of-the-art for various classification tasks, but are computationally expensive. Propagating through the convolutional layers is very slow, as each kernel in each layer must sequentially calculate many inner products for a single forward and backward propagation which equates to O(N^2 n^2) per kernel per layer where the inputs are N x N arrays and the kernels are n x n arrays. Convolution can be efficiently performed as a Hadamard product in the frequency domain. The bottleneck is the transformation which has a cost of O(N^2 log_2 N) using the fast Fourier transform (FFT). However, …

Feature Extraction Using Dimensionality Reduction Techniques: Capturing The Human Perspective, Ashley B. Coleman

Browse all Theses and Dissertations

The purpose of this paper is to determine if any of the four commonly used dimensionality reduction techniques are reliable at extracting the same features that humans perceive as distinguishable features. The four dimensionality reduction techniques that were used in this experiment were Principal Component Analysis (PCA), Multi-Dimensional Scaling (MDS), Isomap and Kernel Principal Component Analysis (KPCA). These four techniques were applied to a dataset of images that consist of five infrared military vehicles. Out of the four techniques three out of the five resulting dimensions of PCA matched a human feature. One out of five dimensions of MDS matched …

Tr-2015001: A Survey And Critique Of Facial Expression Synthesis In Sign Language Animation, Hernisa Kacorri

Computer Science Technical Reports

Sign language animations can lead to better accessibility of information and services for people who are deaf and have low literacy skills in spoken/written languages. Due to the distinct word-order, syntax, and lexicon of the sign language from the spoken/written language, many deaf people find it difficult to comprehend the text on a computer screen or captions on a television. Animated characters performing sign language in a comprehensible way could make this information accessible. Facial expressions and other non-manual components play an important role in the naturalness and understandability of these animations. Their coordination to the manual signs is crucial …

Gender-Based Violence In 140 Characters Or Fewer: A #Bigdata Case Study Of Twitter, Hemant Purohit, Tanvi Banerjee, Andrew Hampton, Valerie L. Shalin, Nayanesh Bhandutia, Amit P. Sheth

Kno.e.sis Publications

Public institutions are increasingly reliant on data from social media sites to measure public attitude and provide timely public engagement. Such reliance includes the exploration of public views on important social issues such as gender-based violence (GBV). In this study, we examine big (social) data consisting of nearly fourteen million tweets collected from Twitter over a period of ten months to analyze public opinion regarding GBV, highlighting the nature of tweeting practices by geographical location and gender. We demonstrate the utility of Computational Social Science to mine insight from the corpus while accounting for the influence of both transient events …

Evaluating Small Drone Surveillance Capabilities To Enhance Traffic Conformance Intelligence, Brian Cusack, Reza Khaleghparast

Australian Security and Intelligence Conference

The availability of cheap small physical drones that fly around and have a variety of visual and sensor networks attached invites investigation for work applications. In this research we assess the capability of a set of commercially available drones (VTOL) that cost less than $1000 (Cheap is a relative term and we consider anything less than $5000 relatively cheap). The assessment reviews the capability to provide secure and safe motor vehicle surveillance for conformance intelligence. The evaluation was conducted by initially estimating a set of requirements that would satisfy an ideal surveillance situation and then by comparing a sample of …

Ieee Access Special Section Editorial: Recent Advances In Software Defined Networking For 5g Networks, Mugen Peng, Tao Huang, Y. Richard Yu, Jianli Pan

Computer Science Faculty Works

No abstract provided.

File System Modelling For Digital Triage: An Inductive Profiling Approach, Benjamin Rice, Benjamin Turnbull

Australian Digital Forensics Conference

Digital Triage is the initial, rapid screening of electronic devices as a precursor to full forensic analysis. Triage has numerous benefits including resource prioritisation, greater involvement of criminal investigators and the rapid provision of initial outcomes. In traditional scientific forensics and criminology, certain behavioural attributes and character traits can be identified and used to construct a case profile to focus an investigation and narrow down a list of suspects. This research introduces the Triage Modelling Tool (TMT), that uses a profiling approach to identify how offenders utilise and structure files through the creation of file system models. Results from the …