Computer Sciences Commons^™

Deep Neural Networks For Multi-Label Text Classification: Application To Coding Electronic Medical Records, Anthony Rios

Theses and Dissertations--Computer Science

Coding Electronic Medical Records (EMRs) with diagnosis and procedure codes is an essential task for billing, secondary data analyses, and monitoring health trends. Both speed and accuracy of coding are critical. While coding errors could lead to more patient-side financial burden and misinterpretation of a patient’s well-being, timely coding is also needed to avoid backlogs and additional costs for the healthcare facility. Therefore, it is necessary to develop automated diagnosis and procedure code recommendation methods that can be used by professional medical coders.

The main difficulty with developing automated EMR coding methods is the nature of the label space. The …

“Woodlands” - A Virtual Reality Serious Game Supporting Learning Of Practical Road Safety Skills., Krzysztof Szczurowski, Matt Smith

Conference Papers

In developed societies road safety skills are taught early and often practiced under the supervision of a parent, providing children with a combination of theoretical and practical knowledge. At some point children will attempt to cross a road unsupervised, at that point in time their safety depends on the effectiveness of their road safety education. To date, various attempts to supplement road safety education with technology were made. Most common approach focus on addressing declarative knowledge, by delivering road safety theory in an engaging fashion. Apart from expanding on text based resources to include instructional videos and animations, some stakeholders …

Implementation Costs Of Spiking Versus Rate-Based Anns, Lacie Renee Stiffler

Theses and Dissertations

Artificial neural networks are an effective machine learning technique for a variety of data sets and domains, but exploiting the inherent parallelism in neural networks requires specialized hardware. Typically, computing the output of each neuron requires many multiplications, evaluation of a transcendental activation function, and transfer of its output to a large number of other neurons. These restrictions become more expensive when internal values are represented with increasingly higher data precision. A spiking neural network eliminates the limitations of typical rate-based neural networks by reducing neuron output and synapse weights to one-bit values, eliminating hardware multipliers, and simplifying the activation …

Improving Speech-Related Facial Action Unit Recognition By Audiovisual Information Fusion, Zibo Meng

Theses and Dissertations

In spite of great progress achieved on posed facial display and controlled image acquisition, performance of facial action unit (AU) recognition degrades significantly for spontaneous facial displays. Furthermore, recognizing AUs accompanied with speech is even more challenging since they are generally activated at a low intensity with subtle facial appearance/geometrical changes during speech, and more importantly, often introduce ambiguity in detecting other co-occurring AUs, e.g., producing non-additive appearance changes. All the current AU recognition systems utilized information extracted only from visual channel. However, sound is highly correlated with visual channel in human communications. Thus, we propose to exploit both audio …

An Analysis Of Software Testing Practices On Migrations From On Premise To Cloud Hosted Environments, Ronan Mullen

Dissertations

This research project examines the differences between software testing practices that are carried out on software that is installed locally (i.e. on premise) versus software that has migrated to a cloud hosted environment. In conjunction with this, focus was placed on determining what methodologies and frameworks are in existence for assisting with software migrations to the cloud. The reason for carrying out this research project was that the transition to cloud computing is becoming more and more mainstream, as a result organisations are required to focus their efforts on how best to move their software to the cloud while ensuring …

Scalable Syriac Paleography Using Interactive Visualization, R. Jordan Crouser, Michael Penn, Nicholas Howe

Computer Science: Faculty Publications

Syriac (a dialect of Aramaic) was the primary language spoken in the late ancient Middle East between the second and eighth centuries AD, and continues to be a language of Christian scholarship and liturgy up to the present day. There are approximately 20,000 known surviving Syriac manuscripts. Among early manuscripts, only around 10% include a scribal note that provides information regarding when, where, and by whom a given manuscript was written. For the remaining 90%, close examination of subtle differences in the handwritten script remains the primary tool for determining provenance. Prior to this study, scholars classified early Syriac manuscripts …

Aligned Sub-Hierarchies: A Structure-Based Approach To The Cover Song Task, Katherine M. Kinnaird

Computer Science: Faculty Publications

Extending previous structure-based approaches to the song comparison tasks such as the fingerprint and cover song tasks, this paper introduces the aligned sub-hierarchies (AsH) representation. Built by applying a post-processing technique to the aligned hierarchies of a song, the AsH representation is the set of unique aligned hierarchies for repeats (called AHR ) encoded in the original aligned hierarchies of the whole song. Effectively each AHR within AsH is a section of the aligned hierarchies for the original song. Like aligned hierarchies, the AsH representation can be embedded into a classification space with a natural metric that makes inter-song comparisons …

Scenario Development For Unmanned Aircraft System Simulation-Based Immersive Experiential Learning, Nickolas D. Macchiarella, Alexander J. Mirot

Journal of Aviation/Aerospace Education & Research

Application of scenario-based training can serve as practical means of educating remote pilots and sensor operators as they seek professional levels of knowledge. Both education and training can build upon time-tested training and simulation methodologies that apply simulators in settings that mirror real-world operations. Embry-Riddle Aeronautical University’s unmanned aircraft system (UAS) program curriculum is rooted in immersive simulation that offers students an experiential learning experience that is aimed to develop higher-order thinking skills. Skills that are critical to professional levels of performance. The degree program builds from basic application skills to critical thinking skills by using immersive scenario-based training in …

Comparing The Effectiveness Of Support Vector Machines And Convolutional Neural Networks For Determining User Intent In Conversational Agents, Kieran O Sullivan

Dissertations

Over the last fifty years, conversational agent systems have evolved in their ability to understand natural language input. In recent years Natural Language Processing (NLP) and Machine Learning (ML) have allowed computer systems to make great strides in the area of natural language understanding. However, little research has been carried out in these areas within the context of conversational systems. This paper identifies Convolutional Neural Network (CNN) and Support Vector Machine (SVM) as the two ML algorithms with the best record of performance in ex isting NLP literature, with CNN indicated as generating the better results of the two. A …

Using Machine Learning Techniques To Predict A Risk Score For New Members Of A Chit Fund Group, Sinead Aherne

Dissertations

Predicting the risk score of new and potential customers is used across the financial industry. By implementing the prediction of risk scores for their customers a chit fund company can improve the knowledge and customer understanding without relying on human knowledge. Data is collected on each customer before they have taken out credit and during the time they contribute to a chit fund. Having collected the necessary data, the company can then decide whether modelling customer risk would benefit them. As the data is available historically, one aspect of risk score prediction will be the focus of this thesis, supervised …

Through The Net: Investigating How User Characteristics Influence Susceptibility To Phishing, Charlie Marriott

Dissertations

In the past 25 years, the internet has grown and evolved from a niche networking technology, used almost exclusively by researchers and enthusiasts, into the driving force of modern economies. Fraud has evolved too, with rates of cybercrime on the increase as criminals become increasingly sophisticated in using technology to deceive their victims. The world is an online place, and data is the new oil. Phishing is a form of social engineering that is not that different from traditional fraud. Phishing attackers try to trick their victims into revealing valuable private information, usually for financial gain, by posing as a …

Automation Of Authorisation Vulnerability Detection In Authenticated Web Applications, Niall Caffrey

Dissertations

In the beginning the World Wide Web, also known as the Internet, consisted mainly of websites. These were essentially information depositories containing static pages, with the flow of information mostly one directional, from the server to the user’s browser. Most of these websites didn’t authenticate users, instead, each user was treated the same, and presented with the same information. A malicious party that gained access to the web server hosting these websites would usually not gain access to confidential information as most of the information on the web server would already be accessible to the public. Instead, the malicious party …

A Demographic Analysis To Determine User Vulnerability Among Several Categories Of Phishing Attacks., Robert Griffin

Dissertations

Phishing attacks have been on a meteoric rise in the last number of years, with 2016 seeing a 65% increase. The attacks range from targeting individuals with personalised messages to spam attacks from bot accounts. With the chances of being targeted by a phishing attack increasing, it is important to identify who is most at risk in order to help alleviate this threat. The aim of this study is to examine members from several demographics and their vulnerability to three types of phishing using data collected from a survey (n = 198). The survey tested the participant’s ability to recognise …

Assesing Completeness Of Solvency And Financial Condition Reports Through The Use Of Machine Learning And Text Classification, Ruairí Nugent

Dissertations

Text mining is a method for extracting useful information from unstructured data through the identification and exploration of large amounts of text. It is a valuable support tool for organisations. It enables a greater understanding and identification of relevant business insights from text. Critically it identifies connections between information within texts that would otherwise go unnoticed. Its application is prevalent in areas such as marketing and political science however, until recently it has been largely overlooked within economics. Central banks are beginning to investigate the benefits of machine learning, sentiment analysis and natural language processing in light of the large …

Augmented Reality As A Potential Tool For Filmmaking, Paul Blachfield

Dissertations

Augmented Reality (AR) has been used for a wide variety of industries. The purpose of this study was to determine the suitability of this technology for use in filmmaking. One of the problems on a film set is the time taken to block a scene. Blocking involves the placement of subjects and props within a scene. Different ideas have been used for blocking including previzualisation and Virtual Reality (VR). This study proposesed the use of AR as a tool to solve this problem. Marker-based and Markerless AR were assessed in turn to determine their suitability for addressing the problem. The …

An Exploration Of Parliamentary Speeches In The Irish Parliament Using Topic Modeling, Fiona Leheny

Dissertations

The only resource available in the public domain which highlights parliamentary ac tivity is parliamentary questions. Up until the last ten years, manual content analysis was carried out to classify these. More recently, machine learning techniques have been used to automatically classify and analyse these data sets. This study analyses the verbal parliamentary speeches in the Irish Parliament (known as the D´ail) over a ten year period using unsupervised machine learning. It does so by applying a less utilised topic modeling technique, known as Non-negative Matrix Factorisation (NMF), to de tect the latent themes in these speeches. A two-layer dynamic …

Intergrating The Fruin Los Into The Multi-Objective Ant Colony System, Tirdad Kiafar

Dissertations

Building evacuation simulation provides the planners and designers an opportunity to analyse the designs and plan a precise, scenario specific instruction for disaster times. Nevertheless, when disaster strikes, the unexpected may happen and many egress paths may get blocked or the conditions of evacuees may not let the execution of emergency plans go smoothly. During disaster times, effective route-finding methods can help efficient evacuation process, in which the directors are able to react to the sudden changes in the environment. This research tries to integrate the highly accepted human dynamics methods proposed by Fruin into the Ant-Colony optimisation route-finding method. …

Identifying And Scoping Context-Specific Use Cases For Blockchain-Enabled Systems In The Wild., Fiona Delaney

Dissertations

Advances in technology often provide a catalyst for digital innovation. Arising from the global banking crisis at the end of the first decade of the 21st Century, decentralised and distributed systems have seen a surge in growth and interest. Blockchain technology, the foundation of the decentralised virtual currency Bitcoin, is one such catalyst. The main component of a blockchain, is its public record of verified, timestamped transactions maintained in an append-only, chain-like, data structure. This record is replicated across n-nodes in a network of co-operating participants. This distribution offers a public proof of transactions verified in the past. Beyond tokens …

Handwritten Digit Recognition And Classification Using Machine Learning, Ke Zhao

Dissertations

In this paper, multiple learning techniques based on Optical character recognition (OCR) for the handwritten digit recognition are examined, and a new accuracy level for recognition of the MNIST dataset is reported. The proposed framework involves three primary parts, image pre-processing, feature extraction and classification. This study strives to improve the recognition accuracy by more than 99% in handwritten digit recognition. As will be seen, pre-processing and feature extraction play crucial roles in this experiment to reach the highest accuracy.

Exploratory Reconstructability Analysis Of Accident Tbi Data, Martin Zwick, Nancy Ann Carney, Rosemary Nettleton

Complex Systems Faculty Publications and Presentations

This paper describes the use of reconstructability analysis to perform a secondary study of traumatic brain injury data from automobile accidents. Neutral searches were done and their results displayed with a hypergraph. Directed searches, using both variable-based and state-based models, were applied to predict performance on two cognitive tests and one neurological test. Very simple state-based models gave large uncertainty reductions for all three DVs and sizeable improvements in percent correct for the two cognitive test DVs which were equally sampled. Conditional probability distributions for these models are easily visualized with simple decision trees. Confounding variables and counter-intuitive findings are …

Ideas And Graphs: The Tetrad Of Activity, Martin Zwick

Complex Systems Faculty Publications and Presentations

A graph can specify the skeletal structure of an idea, onto which meaning can be added by interpreting the structure. This paper considers several directed and undirected graphs consisting of four nodes, and suggests different meanings that can be associated with these different structures. Drawing on John G. Bennett’s “systematics,” specifically on the Tetrad that systematics offers as a model of “activity,” the analysis formalizes and augments the systematics account and shows that the Tetrad is a versatile model of problem-solving, regulation and control, and other processes. Discussion is extended to include hypergraphs, in which links can relate more than …

Hybrid Energy System With Optimized Storage For Improvement Of Sustainability In A Small Town, Fengchang Jiang, Haiyan Xie, Oliver Ellen

Faculty Publications – Technology

With the rise of renewable energy comes significant challenges and benefits. The current studies on the incorporation of renewable-energy policies and energy-storage technologies attempt to address the optimization of hybrid energy systems (HESs). However, there is a gap between the currents needs of HES in small towns for energy independence and the understanding of integrated optimization approaches for employing the technology. The purpose of this research is to determine the technical, systematic and financial requirements needed to allow a city or community to become independent of the utilization of traditional energy and develop a reliable program for a clean and …

Generative Processes For Audification, Judith Jackson

Honors Papers

Using the JavaSerial library, I present a generative method for the digital signal processing technique of audification. By analyzing multiple test instances produced by this method, I demonstrate that a generative audification process can be precise and easily controlled. The parameters tested in this experiment cause explicit, one-to-one changes in the resulting audio of each test instance.

Determinants Of Personal Information Protection Activities In South Korea, Pilku Kang

MPA/MPP/MPFM Capstone Projects

The purpose of this paper is to investigate how people’s awareness and ways to obtain relevant materials of personal information have influenced individual’s information privacy protection activities. This study uses the data of a 2016 survey on information security published by Korea Information and Security Agency.

The dependent variables of this study are preventive measures for the security of a Personal Computer (PC) and preventive measures against personal information breach. I classify independent variables into four types. They are internet users’ perception about information privacy, such as awareness of the importance of protecting one’s personal information, and awareness of information …

Automatic Log Parser To Support Forensic Analysis, Hudan Studiawan, Ferdous Sohel, Christian Payne

Australian Digital Forensics Conference

Event log parsing is a process to split and label each field in a log entry. Existing approaches commonly use regular expressions or parsing rules to extract the fields. However, such techniques are time-consuming as a forensic investigator needs to define a new rule for each log file type. In this paper, we present a tool, namely nerlogparser, to parse the log entries automatically, where log parsing is modeled as a named entity recognition problem. We use a deep machine learning technique, specifically the bidirectional long short-term memory networks, as the underlying architecture for this purpose. Unlike existing tools, nerlogparser …

Detection Techniques In Operational Technology Infrastructure, Glenn Murray, Matthew Peacock, Priya Rabadia, Paresh Kerai

Australian Information Security Management Conference

In previous decades, cyber-attacks have not been considered a threat to critical infrastructure. However, as the Information Technology (IT) and Operational Technology (OT) domains converge, the vulnerability of OT infrastructure is being exploited. Nation-states, cyber criminals and hacktivists are moving to benefit from economic and political gains. The OT network, i.e. Industrial Control System (ICS) is referred to within OT infrastructure as Supervisory Control and Data Acquisition (SCADA). SCADA systems were introduced primarily to optimise the data transfer within OT network infrastructure. The introduction of SCADA can be traced back to the 1960’s, a time where cyber-attacks were not considered. …

Data Visualization And Classification Of Artificially Created Images, Dmytro Dovhalets

All Master's Theses

Visualization of multidimensional data is a long-standing challenge in machine learning and knowledge discovery. A problem arises as soon as 4-dimensions are introduced since we live in a 3-dimensional world. There are methods out there which can visualize multidimensional data, but loss of information and clutter are still a problem. General Line Coordinates (GLC) can losslessly project n-dimensional data in 2- dimensions. A new method is introduced based on GLC called GLC-L. This new method can do interactive visualization, dimension reduction, and supervised learning. One of the applications of GLC-L is transformation of vector data into image data. This novel …

Spike-Based Classification Of Uci Datasets With Multi-Layer Resume-Like Tempotron, Sami Abdul-Wahid

All Master's Theses

Spiking neurons are a class of neuron models that represent information in timed sequences called ``spikes.'' Though predominantly used in neuro-scientific investigations, spiking neural networks (SNN) can be applied to machine learning problems such as classification and regression. SNN are computationally more powerful per neuron than traditional neural networks. Though training time is slow on general purpose computers, spike-based hardware implementations are faster and have shown capability for ultra-low power consumption. Additionally, various SNN training algorithms have achieved comparable performance with the State of the Art on the Fisher Iris dataset. Our main contribution is a software implementation of the …

On Comparability Of Bigrassmannian Permutations, John Engbers, Adam Hammett

Mathematics, Statistics and Computer Science Faculty Research and Publications

Let S_n and G_n denote the respective sets of ordinary and bigrassmannian (BG) permutations of order n, and let (G_n,≤) denote the Bruhat ordering permutation poset. We study the restricted poset (B_n,≤), first providing a simple criterion for comparability. This criterion is used to show that that the poset is connected, to enumerate the saturated chains between elements, and to enumerate the number of maximal elements below r fixed elements. It also quickly produces formulas for β(ω) (α(ω), respectively), the number of BG permutations weakly below (weakly above, respectively) a fixed ω ∈ B …

การทำนายข้อมูลจราจรเชิงพื้นที่และเวลาโดยใช้การฝังข้อมูลอุบัติเหตุร่วมกับนิวรอลเน็ตเวิร์กเชิงลึก, วนิดา ลิยงค์

Chulalongkorn University Theses and Dissertations (Chula ETD)

ระบบขนส่งและจราจรอัจฉริยะ (Intelligent Transportation System, ITS) นั้น มีความสำคัญเป็นอย่างมากต่อการดำรงชีวิตในปัจจุบัน และเมื่อไม่นานมานี้ เริ่มมีการนำการเรียนรู้เชิงลึก (Deep Learning) มาใช้ในการทำนายข้อมูลจราจรเพื่อช่วยให้มีความแม่นยำมากยิ่งขึ้น อย่างไรก็ตาม ปัญหาสำคัญของการทำนายข้อมูลจราจรในเครือข่ายขนาดใหญ่คือการทำนายล่วงหน้าในหลาย ๆ ช่วงเวลา และทำนายในตำแหน่งที่แตกต่างกัน นอกจากนี้สำหรับการจราจรแล้ว อุบัติเหตุที่เกิดขึ้นนั้นจะส่งผลกระทบต่อการจราจรเสมอ การเรียนรู้ถึงผลกระทบที่เกิดขึ้นของอุบัติเหตุจะช่วยให้การทำนายข้อมูลจราจรมีความแม่นยำขึ้น งานวิจัยนี้ จึงนำเสนอนิวรอลเน็ตเวิร์กที่เรียนรู้ความสัมพันธ์ในเชิงพื้นที่และเวลาของข้อมูลจราจร โดยใช้นิวรอลเน็ตเวิร์คแบบคอนโวลูชัน (Convolutional Neural Network, CNN) ร่วมกับหน่วยความจำระยะสั้นแบบยาว (Long Short-Term Memory, LSTM) เพื่อให้สามารถเรียนรู้และทำนายข้อมูลจราจรได้แม่นยำยิ่งขึ้น อีกทั้งยังมีการนำตัวเข้ารหัสอัตโนมัติ (Autoencoder) มาเรียนรู้ข้อมูลอุบัติเหตุ เพื่อให้สามารถเรียนรู้ถึงผลกระทบที่เกิดขึ้นในช่วงที่เกิดอุบัติเหตุไปพร้อม ๆ กันได้