Computer Sciences Commons^™

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.

Towards Dynamic Interaction-Based Reputation Models, Almas Melnikov, Manuel Mazzara, Victor Rivera, Jooyoung Lee, Luca Longo

Articles

In this paper, we investigate how dynamic properties of reputation can influence the quality of users’ ranking. Reputation systems should be based on rules that can guarantee high level of trust and help identify unreliable units. To understand the effectiveness of dynamic properties in the evaluation of reputation, we propose our own model (DIB-RM) that utilizes three factors: forgetting, cumulative, and activity period. In order to evaluate the model, we use data from StackOverflow which also has its own reputation model. We estimate similarity of ratings between DIB-RM and the StackOverflow reputation model to test our hypothesis. We use two …

Pseudorehearsal In Actor-Critic Agents With Neural Network Function Approximation, Vladimir Marochko, Leonard Johard, Manuel Mazzara, Luca Longo

Articles

Catastrophic forgetting has a significant negative impact in reinforcement learning. The purpose of this study is to investigate how pseudorehearsal can change performance of an actor-critic agent with neural-network function approximation. We tested agent in a pole balancing task and compared different pseudorehearsal approaches. We have found that pseudorehearsal can assist learning and decrease forgetting.

Three Decades Of Universal Design - Defining Moments, Margaret Kinsella

Articles

This paper contributes to the growing research on incorporating Universal Design in the Higher Education landscape by presenting a Practitioner's Perspective on Universal Design as delivered in the Institute of Technology, Blanchardstown (ITB) in the first year of a creative digital media degree as part of the first year experience. This first year experience is a transition time for many students and has many complexities; while being an exciting and fulfilling time, the transition can also be challenging and isolating. Through Universal Design, the aim is to enable students to 'Get connected' and 'Stay connected'. Universal Design for Learning is …

A Javascript Framework Comparison Based On Benchmarking Software Metrics And Environment Configuration, Jefferson Ferreira

Dissertations

JavaScript is a client-side programming language that can be used in multi-platform applications. It controls HTML and CSS to manipulate page behaviours and is widely used in most websites over the internet. JavaScript frameworks are structures made to help web developers build web applications faster by offering features that enhance the user interaction with the web page. An increasing number of JavaScript frameworks have been released in recent years in the market to help front-end developers build applications in a shorter space of time. Decision makers in software companies have been struggling to determine which frameworks are best suited for …

Investigation Into The Predictive Power Of Artificial Neural Networks And Logistic Regression For Predicting Default In Chit Funds, Ciara Kerrigan

Dissertations

This study evaluated the performance of an artificial neural network (ANN) multi-layer perceptron model and a logistic regression logitboost (LR) model to predict default in chit funds. The two types of default investigated were late payment of 30 days and late payment of 90 days. The dataset was broken up into training and validation datasets using random sampling and K folds cross validation was used on the training dataset to assess performance of the tuning parameters. The validation dataset was used to compare performance of both algorithms. Principle component analysis (PCA) was used to reduce the feature set while still …

Exploring The Features To Classify The Musical Period Of Western Classical Music, Arturo Martínez Gallardo

Dissertations

Music Information Retrieval (MIR) focuses on extracting meaningful information from music content. MIR is a growing field of research with many applications such as music recommendation systems, fingerprinting, query-by-humming or music genre classification. This study aims to classify the styles of Western classical music, as this has not been explored to a great extent by MIR. In particular, this research will evaluate the impact of different music characteristics on identifying the musical period of Baroque, Classical, Romantic and Modern. In order to easily extract features related to music theory, symbolic representation or music scores were used, instead of audio format. …

Rhythm Inference From Audio Recordings Of Irish Traditional Music, Pierre Beauguitte, Bryan Duggan, John D. Kelleher

Conference papers

A new method is proposed to infer rhythmic information from audio recordings of Irish traditional tunes. The method relies on he repetitive nature of this musical genre. Low-level spectral features and autocorrelation are used to obtain a low-dimensional representation, on which logistic regression models are trained. Two experiments are conducted to predict rhythmic information at different levels of precision. The method is tested on a collec- ion of session recordings, and high accuracy scores are reported.

A new method is proposed to infer rhythmic information from audio recordings of Irish traditional tunes. The method relies on he repetitive nature of …

Validation Of Tagging Suggestion Models For A Hotel Ticketing Corpus, Bojan Bozic, Andre Rios, Sarah Jane Delany

Conference papers

This paper investigates methods for the prediction of tags on a textual corpus that describes hotel staff inputs in a ticketing system. The aim is to improve the tagging process and find the most suitable method for suggesting tags for a new text entry. The paper consists of two parts: (i) exploration of existing sample data, which includes statistical analysis and visualisation of the data to provide an overview, and (ii) evaluation of tag prediction approaches. We have included different approaches from different research fields in order to cover a broad spectrum of possible solutions. As a result, we have …

Non-Linear Machine Learning With Active Sampling For Mox Drift Compensation, Tamara Matthews, Muhammad Iqbal, Horacio Gonzalez-Velez

Conference papers

Abstract—Metal oxide (MOX) gas detectors based on SnO2 provide low-cost solutions for real-time sensing of complex gas mixtures for indoor ambient monitoring. With high sensitivity under ideal conditions, MOX detectors may have poor longterm response accuracy due to environmental factors (humidity and temperature) along with sensor aging, leading to calibration drifts. Finding a simple and efficient solution to correct such calibration drifts has been the subject of numerous studies but remains an open problem. In this work, we present an efficient approach to MOX calibration using active and transfer sampling techniques coupled with non-linear machine learning algorithms, namely neural networks, …

The Terror Network Industrial Complex: A Measurement And Analysis Of Terrorist Networks And War Stocks, James Usher, Pierpaolo Dondio

Conference papers

This paper presents a measurement study and analysis of the structure of multiple Islamic terrorist networks to determine if similar characteristics exist between those networks. We examine data gathered from four terrorist groups: Al-Qaeda, ISIS, Lashkar-e-Taiba (LeT) and Jemaah Islamiyah (JI) consisting of six terror networks. Our study contains 471 terrorists’ nodes and 2078 links. Each terror network is compared in terms efficiency, communication and composition of network metrics. The paper examines the effects these terrorist attacks had on US aerospace and defence stocks (herein War stocks). We found that the Islamic terror groups increase recruitment during the planned attacks, …

A Proposal To Embed The In Dubio Pro Reo Principle Into Abstract Argumentation Semantics Based On Topological Ordering And Undecidedness Propagation, Pierpaolo Dondio, Luca Longo

Conference papers

Abstract. In this paper we discuss how the in dubio pro reo principle and the corresponding standard of proof beyond reasonable doubt can be modelled in abstract argumentation. The in dubio pro reo principle protects arguments against attacks from doubtful arguments. We identify doubtful arguments with a subset of undecided arguments, called active undecided arguments, consisting of cyclic arguments responsible for generating the undecided situation. We obtain the standard of proof beyond reasonable doubt by imposing that attacks from doubtful undecided arguments are not enough to change the acceptability status of an attacked argument (the reo). The resulting semantics, called …

A Comparison Of Classical Versus Deep Learning Techniques For Abusive Content Detection On Social Media Sites, Hao Che, Susan Mckeever, Sarah Jane Delany

Conference papers

The automated detection of abusive content on social media websites faces a variety of challenges including imbalanced training sets, the identification of an appropriate feature representation and the selection of optimal classifiers. Classifiers such as support vector machines (SVM), combined with bag of words or ngram feature representation, have traditionally dominated in text classification for decades. With the recent emergence of deep learning and word embeddings, an increasing number of researchers have started to focus on deep neural networks. In this paper, our aim is to explore cutting-edge techniques in automated abusive content detection. We use two deep learning approaches: …

Predicting Happiness - Comparison Of Supervised Machine Learning Techniques Performance On A Multiclass Classification Problem, Dorota Nieciecka

Dissertations

In the modern world, especially in contemporary economies and politics, a population's subjective well-being is a frequent subject of the public debate. As comparisons of happiness levels in different countries are published, different circumstances and their effect on the value of the subjective well-being reported by people are also analysed. However, a significant amount of the research related to subjective well-being and its determinants is still based upon survey answers and employing conventional statistical methods providing details regarding correlations and causality between different factors and subjective well-being. Application of Supervised Machine Learning techniques for prediction of subjective well-being may provide …

An Investigation Of The Impact Of A Social Constructivist Teaching Approach, Based On Trigger Questions, Through Measures Of Mental Workload And Efficiency, Federico Gobbo, Luca Longo, Declan O'Sullivan, Giuliano Orru

Conference papers

Social constructivism is grounded on the construction of information with a focus on collaborative learning through social interactions. However, it tends to ignore the human mental architecture, pillar of cognitivism. A characteristic of cognitivism is that instructional designs built upon it are generally explicit, contrarily to constructivism. This position paper proposes a novel learning task that is aimed at combining both the approaches through the use of trigger questions in a collaborative activity executed after a traditional delivery of instructions. To evaluate this new task, a metric of efficiency based upon a measure of mental workload and a measure of …

Image Classification Using Bag-Of-Visual-Words Model, Kaiqiang Huang

Dissertations

Recently, with the explosive growth of digital technologies, there has been a rapid proliferation of the size of image collection. The technique of supervised image clas sification has been widely applied in many domains in order to organize, search, and retrieve images. However, the traditional feature extraction approaches yield the poor classification accuracy. Therefore, the Bag-of-visual-words model, inspired by Bag-of Words model in document classification, was used to present images with the local descriptors for image classification, and also it performs well in some fields. This research provides the empirical evidence to prove that the BoVW model outperforms the traditional …

That Seems Made Up: Deep Learning Classifiers For Fiction & Non Fiction Book Reviews, Clement Manger

Dissertations

The thesis aims to take the first step towards automated extraction of the information found in book reviews, by using machine learning tools to assign a label of fiction or non fiction to the text. The thesis makes use of neural networks and performs experiments around architecture, hyper-parameters and text processing from which an optimized model is produced. The thesis enjoys certain successes; it was possible to match the state of the art achieved by (Kim, 2014) and computation was sped up considerably from the default to the optimized model by 13.8 seconds per 50 steps. Further it is confirmed …

Hollow Core Fiber Based Interferometer For High Temperature (1000 °C) Measurement, Dejun Liu, Qiang Wu, Chao Mei, Jinhui Yuan, Xiangjun Xin, Arun Mallik, Fangfang Wei, Wei Han, Rahul Kumar, Chongxiu Yu, Shengpeng Wan, Xingdao He, Bo Liu, Gang-Ding Peng, Yuliya Semenova, Gerald Farrell

Articles

A simple, cost effective high temperature sensor (up to 1000 °C) based on a hollow core fiber (HCF) structure is reported. It is configured by fusion splicing a short section of HCF with a length of few millimeters between two standard single mode fibers (SMF-28). Due to multiple beam interference introduced by the cladding of the HCF, periodic transmission dips with high spectral extinction ratio and high quality (Q) factor are excited. However, theoretical analysis shows that minor variations of the HCF cladding diameter may result in a significant decrease in the Q factor. Experimental results demonstrate that the position …

Sensory Seduction And Narrative Pull, Nina Lyons, Matt Smith, Hugh Mccabe

Conference papers

User experience design is a process that has been defined, developed and refined over the last few decades. It is a process of shaping a user's movements through a website or mobile application. It is user-focussed, prioritising utility, ease-of-use and efficiency. It is widely used and has helped advance the way in which users interact with websites and mobile applications, making it far less frustrating. User experience design is a key element in how the internet and mobile technology have become ubiquitous in our daily lives. Given this success, it would seem that continuing to use this process for new …

Modular Mechanistic Networks: On Bridging Mechanistic And Phenomenological Models With Deep Neural Networks In Natural Language Processing, Simon Dobnik, John D. Kelleher

Books/Book chapters

Natural language processing (NLP) can be done using either top-down (theory driven) and bottom-up (data driven) approaches, which we call mechanistic and phenomenological respectively. The approaches are frequently considered to stand in opposition to each other. Examining some recent approaches in deep learning we argue that deep neural networks incorporate both perspectives and, furthermore, that leveraging this aspect of deep learning may help in solving complex problems within language technology, such as modelling language and perception in the domain of spatial cognition.

What Is Not Where: The Challenge Of Integrating Spatial Representations Into Deep Learning Architectures, John D. Kelleher, Simon Dobnik

Books/Book chapters

This paper examines to what degree current deep learning architectures for image caption generation capture spatial lan- guage. On the basis of the evaluation of examples of generated captions from the literature we argue that systems capture what objects are in the image data but not where these objects are located: the cap- tions generated by these systems are the output of a language model conditioned on the output of an object detector that cannot capture fine-grained location information. Although language models provide useful knowledge for image captions, we argue that deep learning image captioning architectures should also model geometric …

On Demonstrating The Impact Of Defeasible Reasoning Via A Multi-Layer Argument-Based Framework (Doctoral Consortium), Lucas Middeldorf Rizzo

Conference papers

Promising results have indicated Argumentation Theory as a solid research area for implementing defeasible reasoning in practice. However, applications are usually domain dependent, not incorporating all the layers and steps required in an argumentation process, thus limit- ing their applicability in different areas. This PhD project is focused on the development of a multi-layer defeasible argument-based framework which is in turn used across different applications in the fields of decision making and knowledge representation and reasoning. The inference produced is compared against the inference of different quantitative theories of reasoning under uncertainty such as expert systems and fuzzy logic. The …

Streaming Vr For Immersion: Quality Aspects Of Compressed Spatial Audio, Miroslaw Narbutt, Sean O’Leary, Andrew Allen, Jan Skoglund, Andrew Hines

Conference papers

Delivering a 360-degree soundscape that matches full sphere visuals is an essential aspect of immersive VR. Ambisonics is a full sphere surround sound technique that takes into account the azimuth and elevation of sound sources, portraying source location above and below as well as around the horizontal plane of the listener. In contrast to channel-based methods, ambisonics representation offers the advantage of being independent of a specific loudspeaker set-up. Streaming ambisonics over networks requires efficient encoding techniques that compress the raw audio content without compromising quality of experience (QoE). This work investigates the effect of audio channel compression via the …

Rating By Ranking: An Improved Scale For Judgement-Based Labels, Jack O'Neill, Sarah Jane Delany, Brian Mac Namee

Conference papers

Labels representing value judgements are commonly elicited using an interval scale of absolute values. Data collected in such a manner is not always reliable. Psychologists have long recognized a number of biases to which many human raters are prone, and which result in disagreement among raters as to the true gold standard rating of any particular object. We hypothesize that the issues arising from rater bias may be mitigated by treating the data received as an ordered set of preferences rather than a collection of absolute values. We experiment on real-world and artificially generated data, finding that treating label ratings …

A Case Study For Ecampus Spatial: Business Data Exploration, James Carswell, Thanh Thao Pham Ti, Andrea Ballatore, Junjun Yin, Linh Truong-Hong

Books/Book chapters

Location based querying is the core interaction paradigm between mobile citizens and the Internet of Things, so providing users with intelligent web-services that interact efficiently with web and wireless devices to recommend personalised services is a key goal. With today's popular Web Map Services, users can ask for general information at a specific location, but not detailed information such as related functionality or environments. This shortcoming comes from a lack of connection between non-spatial “business” data and spatial “map” data. This chapter presents a novel approach for location-based querying in web and wireless environments, in which non-spatial business data is …