Course Overview

Artificial Intelligence is one the most exciting and fastest growing areas of the ICT industry and research. It has the potential to positively transform every aspect of all our lives, from Smart Cities and Autonomous Vehicles, through to improved healthcare services and low-carbon economies. Artificial Intelligence has the capacity to provide intelligent solutions that can help us to tackle many today’s greatest societal challenges. 

Become part of this exciting development by joining our one-year MSc in Computer Science—Artificial Intelligence. 

This programme is aimed at graduates with a primary qualification in Computer Science or related subject area. It is not a conversion course, but expects students to be already at a very high technical standard with regard to their Computer Science education. 

The full-time MSc in Artificial Intelligence is taught by renowned, interdisciplinary NUI Galway experts in the field. It covers over two semesters many complementary areas of Artificial Intelligence, including Meta-Heuristic Optimisation, Deep Learning, Autonomous Agents and Multi-Agent Systems. 

Furthermore, students reinforce their newly gained skills in a 30-credit project that is completed during the summer. Here students may have the opportunity to collaborate with a research group or with an industry partner.

Scholarships available
Find out about our Postgraduate Scholarships here.

Applications and Selections

Applications are made online via the NUI Galway Postgraduate Applications System

Your online application shall include: 

  • A personal statement outlining:
    • A summary of your primary degree and its relevance for a successful completion of this programme. We strongly encourage an evidence-based approach to highlighting your academic accomplishments.
    • A summary of your previous capstone projects (e.g. undergraduate final year projects) including an outline of your exact contribution to these. We strongly encourage an evidence-based approach to outlining your existing technical skills and experience.
  • Your CV
  • University Degree Transcripts
  • Two references
  • IELTS/TOEFL certificate—only if English is not your mother tongue 

What is not required (please do not submit these):

  • Secondary school certificates
  • Training certificates
  • Membership certificates

Who Teaches this Course

  • Professor Michael Madden
  • Dr Enda Barrett
  • Dr Attracta Brennan
  • Dr Paul Buitelaar
  • Dr Des Chambers
  • Dr Jim Duggan
  • Dr Frank Glavin
  • Dr Josephine Griffith
  • Dr Conor Hayes
  • Dr Seamus Hill
  • Dr Enda Howley
  • Dr Eddie Jones
  • Dr John McCrae
  • Dr James McDermott
  • Dr Hugh Melvin
  • Dr Owen Molloy
  • Dr Conn Mulvihill
  • Dr Matthias Nickles
  • Dr Colm O'Riordan
  • Dr David O'Sullivan
  • Dr Sam Redfern
  • Dr Heike Schmidt-Felzmann
  • Dr Michael Schukat
  • Dr Finlay Smith
  • Karen Young

Requirements and Assessment

Key Facts

Entry Requirements

Prior Qualification:
This MSc is targeted at high-performing graduates of level 8 computer science programmes, or level 8 science / engineering programmes that offer sufficient training in computing.

The minimum academic requirement for entry to the programme is a First Class Honours (or equivalent) from a recognised university or third-level college. However, a good Second Class Honours (or equivalent) can be deemed sufficient on the recommendation of the Programme Director. 

English Language Proficiency:
Overall, entry to the MSc in Computer Science—Artificial Intelligence requires a minimum IELTS score of 6.5, with no less than 6.5 in the writing ability category and no less than 6.0 in the other categories.


Additional Requirements

Duration

12 months, full-time

Next start date

September 2020

A Level Grades ()

Average intake

20

Closing Date

 Please view the offer rounds website.

NFQ level

Mode of study

ECTS weighting

90

Award

CAO

Course code

1MAI1

Course Outline

The MSc in Computer Science—Artificial Intelligence is a one-year 90-ECTS course with three main elements:

  • foundational modules (35 ECTS)
  • advanced modules (25 ECTS), and
  • a substantial capstone project (30 ECTS).

Foundational modules include: Machine Learning and Deep Learning; Natural Language Processing; Information Retrieval; Meta-Heuristic Optimisation; Ethics in Artificial Intelligence; Autonomous Agents and Multi-Agent Systems. 

Advanced modules include: Programming and Tools for AI; Knowledge Representation & Statistical Relational Learning; Data Visualisation; Programming for Data Analytics; Web & Network Science; Embedded Image Processing; Tools and Techniques for Large Scale Data Analytics; Research Topics in AI. 

From Semester II onwards, students work on individual projects and submit them in August. Projects may have a research or applied focus.

Curriculum Information

Curriculum information relates to the current academic year (in most cases).
Course and module offerings and details may be subject to change.

Glossary of Terms

Credits
You must earn a defined number of credits (aka ECTS) to complete each year of your course. You do this by taking all of its required modules as well as the correct number of optional modules to obtain that year's total number of credits.
Module
An examinable portion of a subject or course, for which you attend lectures and/or tutorials and carry out assignments. E.g. Algebra and Calculus could be modules within the subject Mathematics. Each module has a unique module code eg. MA140.
Optional
A module you may choose to study.
Required
A module that you must study if you choose this course (or subject).
Semester
Most courses have 2 semesters (aka terms) per year.

Year 1 (90 Credits)

Required CT4100: Information Retrieval


Semester 1 | Credits: 5

The course introduces some of the main theories and techniques in the domain of information retrieval.
(Language of instruction: English)

Learning Outcomes
  1. Explain the main models used information retrieval.
  2. Explain the factors involved in designing and analysing weighting schemes
  3. Be able to chose suitable data structures and algorithms for builing IR systems
  4. Be able to explain the main ideas and approaches used web search, collaborative filtering, multimedia IR
Assessments
  • Written Assessment (70%)
  • Continuous Assessment (30%)
Module Director
Lecturers / Tutors
The above information outlines module CT4100: "Information Retrieval " and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Required CT475: Machine Learning & Data Mining


Semester 1 | Credits: 5

Definitions of Machine Learning, Data Mining and the relationship between them; the CRISP Data Mining process model; major tasks including classification, regression, clustering, association learning, feature selection, and reinforcement learning; algorithms for these tasks that may include decision tree learning, instance-based learning, probabilistic learning, support vector machines, linear and logistic regression, and Q-learning; open-source software tools for data mining; practical applications such as sensor data analysis, healthcare data analysis, and text mining to identify spam email; ethical issues and emerging trends in data mining and machine learning.
(Language of instruction: English)

Learning Outcomes
  1. Define Machine Learning and Data Mining and discuss their relationship
  2. Explain what major categories of Machine Learning tasks entail
  3. Demonstrate how to apply the Data Mining process to practical problems
  4. Explain and apply algorithms for decision tree learning, instance-based learning, linear and logistic regression, probabilistic learning, support vector machines, and reinforcement learning
  5. Given a dataset and data mining task to be addressed, select, apply and evaluate appropriate algorithms, and interpret the results
  6. Discuss ethical issues and emerging trends in data mining and machine learning.
Assessments
  • Written Assessment (75%)
Module Director
Lecturers / Tutors
The above information outlines module CT475: "Machine Learning & Data Mining" and is valid from 2015 onwards.
Note: Module offerings and details may be subject to change.

Required CT5142: Artificial Intelligence and Ethics


Semester 1 | Credits: 5

Artifical intelligence technologies have evolved dramatically in recent years, impacting on many areas of human life. Societal responses to these developments have ranged from enthusiastic optimism to deep suspicion. The module will explore prominent ethical issues arising in relation to the design, use and societal impact of Artificial Intelligence. Topics addressed in the module include Philosophy of Technology, Value Sensitive Design, Responsible Research and Innovation (RRI), Privacy and consent, Contextual integrity, Transparency and explainable AI, Trust and Trustworthiness, Datafication, Algorithmic surveillance, Algorithmic Bias, Autonomous artificial agents and responsibility, and Human replacement.
(Language of instruction: English)

Learning Outcomes
  1. Identify and summarise important ethical concerns related to the design, use and societal impact of Artificial Intelligence.
  2. Apply relevant theoretical models from the ethical, legal and social science literature to identified ethical concerns regarding AI.
  3. Critically analyse strengths and weaknesses of different positions from the ethical, legal and social science literature on ethical concerns related to the design, use and societal impact of Artifical Intelligence.
  4. Demonstrate the ability to communicate core insights from divergent perspectives on ethical concerns coherently and concisely.
Assessments
  • Continuous Assessment (100%)
Module Director
Lecturers / Tutors
Reading List
  1. "Towards a Code of Ethics for Artificial Intelligence" by Paula Boddington
    Publisher: Springer
  2. "Privacy in Context: Technology, Policy, and the Integrity of Social Life" by Helen Nissenbaum
    Publisher: Stanford University Press
  3. "Privacy as Trust: Information Privacy for an Information Age" by Ari Waldman
    Publisher: Cambridge University Press
  4. "Privacy, Big Data, and the Public Good: Frameworks for Engagement" by Julia Lane, Victoria Stodden, Stefan Bender, Helen Nissenbaum (Editors)
    Publisher: Cambridge University Press
  5. "Robot Ethics 2.0: From Autonomous Cars to Artificial Intelligence" by Patrick Lin, Keith Abney, Ryan Jenkins (Editors)
    Publisher: Oxford University Press
  6. "Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor" by Virginia Eubanks
    Publisher: St Martin's Press
The above information outlines module CT5142: "Artificial Intelligence and Ethics" and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Required CT5132: Programming and Tools for AI


Semester 1 | Credits: 5

This module is about programming and computational tools required for artificial intelligence. It uses the Python language as the main vehicle, but focusses on conceptual material rather than just the language itself. It moves fast through introductory Python workings. It covers several important Python libraries in detail. It discusses approaches to building re-usable, high quality code but not on software engineering. It also visits some extra topics such as version control and introduction to the R language for statistics. The module is core for the NUI Galway MSc in Artificial Intelligence (MScAI) Part-time (online) and Full-time (classroom). The syllabus and assessment will be the same for both. We will use a hybrid of lecture-style and lab-style delivery. The lecture-style delivery will be via video (for the part-time/online version) or classroom (for the full-time version). Practical exercises will be interleaved with the lecture-style delivery. This module will be divided into the following main topics: 1. Weeks 1-2: Introductory Python: writing and executing Python code through an IDE, command line, or notebook; arithmetic; syntax; comments and doc-strings; variables; functions; loops and conditionals; lists, tuples, dictionaries; classes; input/output; version control. 2. Weeks 3-4: Python data libraries: Numpy, Pandas, Matplotlib, and friends. 3. Weeks 5-6: Introductory R: some side-by-side comparisons between R and Python; R for statistics. 4. Weeks 7-11: Python software for AI: Scikit-learn API (but not details of the algorithms themselves), NetworkX, and many examples. 5. Week 12: Testing, notebooks, cloud execution. We will use up-to-date versions of software, and in particular we will use Python 3 (not Python 2).
(Language of instruction: English)

Learning Outcomes
  1. Read and write simple Python programs, e.g. for data munging, with a high degree of comfort.
  2. Use R for simple statistics and data exploration.
  3. Use numerical Python libraries for manipulation, input/output, visualisation of numerical data using Numpy array types.
  4. Use essential tools for AI, including regular expressions for text processing; libraries for data gathering, machine learning, combinatorial programming, and modelling networks.
  5. Plan/design a program using any of the above facilities; test it; document it; execute it locally or in the cloud as appropriate.
Assessments
  • Written Assessment (50%)
  • Continuous Assessment (50%)
Module Director
Lecturers / Tutors
The above information outlines module CT5132: "Programming and Tools for AI" and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Required CT5141: Optimisation


Semester 1 | Credits: 5

This module covers optimisation -- "the science of better". Optimisation is used in a huge variety of applications, including: finding time-saving transport routes; scheduling exams without conflicts; reducing weight and cost in engineering design; designing portfolios of financial investments; finding numerical data models with low expected error; and many more. In this module we will aim to understanding a broad range of applications and a unifying view of the field, and concentrate on two main types of methods: (1) metaheuristic optimisation and (2) exact methods for constrained optimisation. In this module we will not cover gradient descent and related methods, as they are covered in machine learning modules available on the MScAI. We will spend time in-class on practical implementations, writing our own optimisation programs from scratch and also using state-of-the-art libraries.
(Language of instruction: English)

Learning Outcomes
  1. Design and implement a variety of metaheuristic algorithms applicable to a variety of domains
  2. Understand objective functions, local/global optima and objective function landscapes
  3. Understand Pareto efficiency, frontiers and dominance
  4. Gain practical knowledge of multiple approaches to optimisation
  5. Explain how to choose one type of algorithm (exact methods, heuristic methods, multi-objective) and one representation over another for a given problem
  6. Implement customised problem-specific algorithms
Assessments
  • Written Assessment (70%)
  • Continuous Assessment (30%)
Module Director
Lecturers / Tutors
The above information outlines module CT5141: "Optimisation" and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Required CT5120: Introduction to Natural Language Processing


Semester 1 | Credits: 5

Introduction to natural language processing, including foundations in linguistics, statistical analysis and applications.
(Language of instruction: English)

Learning Outcomes
  1. Understand the various levels of linguistic structure relevant to NLP.
  2. Use and understand standard algorithms for basic NLP analysis
  3. Gain practical knowledge of and experience in the use of NLP toolkits
  4. Understand the theoretical principles behind core NLP applications.
  5. Apply NLP algorithms, toolkits and applications to Data Analytics tasks.
Assessments
  • Written Assessment (50%)
  • Continuous Assessment (50%)
Module Director
Lecturers / Tutors
The above information outlines module CT5120: "Introduction to Natural Language Processing" and is valid from 2018 onwards.
Note: Module offerings and details may be subject to change.

Required CT5129: Artificial Intelligence Project


15 months long | Credits: 30

This project tests the students ability to carry out in depth analysis, problem-solving and reporting of an AI problem.
(Language of instruction: English)

Learning Outcomes
  1. Apply a variety of artificial intelligence techniques to solve a real world problem.
  2. Diagnose a problem and design an AI based solution.
  3. Conduct and report on exploratory analysis of the problem domain
  4. Produce an in-depth report (thesis) describing the problem, the diagnosis and approaches to solving it.
  5. Demonstrate that they can research, apply and evaluate state-of-the-art techniques in AI.
Assessments
  • Research (100%)
Module Director
Lecturers / Tutors
The above information outlines module CT5129: "Artificial Intelligence Project" and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Required CT5135: Research Topics in AI


Semester 2 | Credits: 5

Research Topics in AI will provide an in-depth coverage of two or three active research areas in the field of AI. For each topic the module provides an overview of research in the area and the implications from real-world implementations.
(Language of instruction: English)

Learning Outcomes
  1. Gain both a wide and a deep knowledge of the topic(s) in the current offering of the module.
  2. Improve their skills at navigating through, and critically examining, the scientific literature on the selected topic(s).
  3. Demonstrate the use of techniques from the selected topic(s) using real-world datasets and tools.
Assessments
  • Continuous Assessment (100%)
Module Director
Lecturers / Tutors
The above information outlines module CT5135: "Research Topics in AI" and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Required CT5100: Data Visualisation


Semester 2 | Credits: 5

This module with teach the fundamentals of data visualization. It will cover basic design principles and the principles underlying human perception, color theory and narrative. It will focus on the use of open standards for the presentation of data on the Web such as HTML, CSS, SVG, javascript through the use of libraries such as D3.js, jQuery.js and Dimple.js.
(Language of instruction: English)

Learning Outcomes
  1. be able to describe the basic design principles underlying human perception, color theory and narrative
  2. demonstrate how to select different types of visualisations to illustrate concepts and data types
  3. know how to use the grammar of graphics to build interpretable static and interactive visualisations using R's visualisation packages
  4. deploy visualisation techniques during exploratory analysis work flow
  5. know how to deploy visualisation libraries such as D3.js and dimple.js
Assessments
  • Continuous Assessment (100%)
Module Director
Lecturers / Tutors
Reading List
  1. "Interactive Data Visualization for the Web" by Scott Murray
    ISBN: 9781449339739.
    Publisher: O'Reilly Media
The above information outlines module CT5100: "Data Visualisation" and is valid from 2018 onwards.
Note: Module offerings and details may be subject to change.

Required CT5134: Agents, Multi-Agent Systems and Reinforcement Learning


Semester 2 | Credits: 5

The topic of Agents and Multi-Agent Systems, examines environment that involve autonomous decision making software actors to interact with their surroundings with the aim of achieving some individual or overall goal. A typical agent environment could be a trading environment where an agent attempts to optimise energy usage, or the profitability of a transaction. More recently, significant global attention has focussed on the vision of autonomous vehicles, which also follows the core principle of an agent attempting to achieve a set of defined goals. This module begins by examining the underpinnings of what is an Agent, and how we can better understand the principles of an agent and its autonomy. Multi-Agent Systems are then explored, as a means of understanding how many agents can interact with each other in a complex environment. Agents are commonly modelled using Game Theory, and in this module a range of Game Theoretic Models will be studied. The module will examine Adaptive Learning Agents through the use of Reinforcement Learning algorithms an area of Machine Learning, which focuses on training learners to choose actions which yield the maximum reward in the absence of prior knowledge. The module takes a hands-on, practical approach to reinforcement learning theory, beginning with Markov Decision Processes, detailing practical learning examples in discrete environments and how to formulate a reinforcement learning task. It then extends this to continuous problem spaces, detailing Deep Reinforcement Learning with a practical implementation of a Deep Q Network using Keras.
(Language of instruction: English)

Learning Outcomes
  1. Explain and discuss the principles underlying Agents.
  2. Explain the role of game theory and games in agent design.
  3. Apply the principle of agents to a range of simulation problems.
  4. Understand the theory unpinning reinforcement learning.
  5. Apply reinforcement learning to a real-world problem.
  6. Apply advanced deep reinforcement learning approaches to a real-world problem.
Assessments
  • Written Assessment (70%)
  • Continuous Assessment (30%)
Module Director
Lecturers / Tutors
The above information outlines module CT5134: "Agents, Multi-Agent Systems and Reinforcement Learning " and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Required CT5133: Deep Learning


Semester 2 | Credits: 5

This is an advanced module in machine learning, focusing on neural networks (NNs), deep NNs, and connectionist computing. Students learn about the basic principles and building blocks of deep learning, and how to implement a deep neural network ‘from scratch’. They also learning about software libraries and tools, and gain experience of applying deep learning in a range of practical applications. The module includes substantial practical programming assignments. This module is intended for students who have completed a first course in machine learning, and already have a good grounding in supervised learning topics including: classification and regression; evaluation of classifiers; overfitting and underfitting; basic algorithms such as k-nearest neighbours, decision tree learning, logistic regression, and gradient descent.
(Language of instruction: English)

Learning Outcomes
  1. Discuss and apply feature engineering and ensembles to improve the performance of classification and regression algorithms, and explain their relationship to connectionist computing
  2. Explain the operation of feed-forward neural networks and the back-propagation algorithm
  3. Describe, implement and apply key features of deep learning
  4. Implement NNs for supervised machine learning tasks, from first principles and (separately) using modern libraries and frameworks
  5. Diagnose common NN problems such as overfitting and underfitting and propose solutions
  6. Choose, explain and implement convnet design for image processing tasks
  7. Choose, explain and implement advanced architectures for specialised tasks
  8. Discuss ethical issues, limitations, and emerging trends in deep learning.
Assessments
  • Written Assessment (70%)
  • Continuous Assessment (30%)
Module Director
Lecturers / Tutors
The above information outlines module CT5133: "Deep Learning" and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Optional EE445: Digital Signal Processing


Semester 1 | Credits: 5

Syllabus Outline: Discrete-time systems, time-domain analysis. The z-Transform. Frequency-domain analysis, the Fourier Transform. Digital filter structures and implementation. Spectral analysis with the DFT, practical considerations. Digital filter design: IIR, FIR, window methods, use of analogue prototypes.
(Language of instruction: English)

Learning Outcomes
  1. Analyse a discrete-time system through calculation of its time-domain properties; in particular, calculate its impulse response, or the system output to any arbitrary input signal.
  2. Describe signals and systems in terms of their z-transforms, and use appropriate techniques to analyse and manipulate them.
  3. Determine the characteristics of a signal or system in the frequency domain, by means of the Fourier Transform, and determine the frequency content in the signal.
  4. Given a discrete-time system description, determine an appropriate structure for implementation (e.g. cascade, parallel), and carry out system design.
  5. Analyse and design specialised digital filters, including notch filters, resonators and oscillators.
  6. Choose appropriate parameters for spectral analysis using the DFT, across a number of applications.
  7. Analyse the computational requirments of time-domain and frequency-domain approaches to implementing digital filters.
  8. Given a required digital filter specification, choose an appropriate design procedure from a number of alternatives, carry out this procedure to determine the required filter transfer function, and verify that the specification has been met.
Assessments
  • Written Assessment (80%)
  • Continuous Assessment (20%)
Module Director
Lecturers / Tutors
The above information outlines module EE445: "Digital Signal Processing" and is valid from 2015 onwards.
Note: Module offerings and details may be subject to change.

Optional CT5105: Tools and Techniques for Large Scale Data Analytics


Semester 1 | Credits: 5

Large-scale data analytics is concerned with the processing and analysis of large quantities of data, typically from distributed sources (such as data streams on the internet). This module introduces students to state-of-the-art approaches to large-scale data analytics. Students learn about foundational concepts, software tools and advanced programming techniques for the scalable storage, processing and predictive analysis of high- volume and high-velocity data, and how to apply them to practical problems. <p><p> ** This module uses Java as programming language. Knowledge of Java is a prerequisite for participation in this module. ** <p><p> Planned topics include: Definition of large-scale computational data analytics; Overview of approaches to the processing and analysis of high volume and high velocity data from distributed sources; Applications of large-scale data analytics; Foundations of cluster computing and parallel data processing; The Hadoop and Spark ecosystems. MapReduce; Advanced programming concepts for large-scale data analytics; Concepts and tools for large-scale data storage; Stream data analytics. Complex Event Processing (CEP); Techniques and open-source tools for large-scale predictive analytics; Computational statistics and machine learning with large-scale data processing frameworks such as Spark; Privacy in the context of large-scale data analytics.
(Language of instruction: English)

Learning Outcomes
  1. Be able to define large-scale data analytics and understand its characteristics
  2. Be able to explain and apply concepts and tools for distributed and parallel processing of large-scale data
  3. Know how to explain and apply concepts and tools for highly scalable collection, querying, filtering, sorting and synthesizing of data
  4. Know how to describe and apply selected statistical and machine learning techniques and tools for the analysis of large-scale data
  5. Know how to explain and apply approaches to stream data analytics and complex event processing
  6. Understand and be able to discuss privacy issues in connection with large-scale data analytics
Assessments
  • Written Assessment (70%)
  • Continuous Assessment (30%)
Module Director
Lecturers / Tutors
Reading List
  1. "Learning Spark: Lightning-Fast Big Data Analytics." by Holden Karau, Andy Konwinski, Patrick Wendell, Matei Zaharia
    Publisher: O'Reilly
  2. "Hadoop: The Definitive Guide" by Tom White
    ISBN: 9781449311520.
    Publisher: O'Reilly Media
  3. "Large-Scale Data Analytics" by Gkoulalas-Divanis, Aris, Labbi, Abderrahim (Eds.)
    ISBN: 1461492424.
    Publisher: Springer
The above information outlines module CT5105: "Tools and Techniques for Large Scale Data Analytics" and is valid from 2018 onwards.
Note: Module offerings and details may be subject to change.

Optional CT561: Systems Modelling and Simulation


Semester 1 | Credits: 5

Simulation is a quantitative method used to support decision making and predicting system behaviour over time. This course focuses the system dynamics approach. The course covers the fundamentals of simulation, and describes how to design and build mathematical models. Case studies used include: software project management, public health policy planning, and capacity planning.
(Language of instruction: English)

Learning Outcomes
  1. Define the aim of Simulation and its role in the decision making process for complex systems
  2. Distinguish between the two feedback types: positive and negative
  3. Demonstrate how to apply the system dynamics approach to areas including public health, software engineering management and capacity planning.
  4. Explain and apply numerical integration methods to solve simulation problems.
  5. Given a simulation problem, formulate a model, test the structure and equations, and perform detailed sensitivity analysis on the impact of a range of policy options
  6. Build, test and evaluate models using Vensim and R.
  7. Appreciate the diiferences between continuous, discrete event and agent-based simulation
Assessments
  • Written Assessment (100%)
Module Director
Lecturers / Tutors
The above information outlines module CT561: "Systems Modelling and Simulation" and is valid from 2016 onwards.
Note: Module offerings and details may be subject to change.

Optional CT5121: Advanced Topics in Natural Language Processing


Semester 2 | Credits: 5

Advanced topics in natural language processing, including deep learning for NLP, machine translation and language resources.
(Language of instruction: English)

Learning Outcomes
  1. Understand vector models of text and be able to apply them.
  2. Know principles of machine translation and be able to build simple machine translation systems.
  3. Gain knowledge of use of deep learning for NLP with applications to machine translation.
  4. Apply data science principles to the use of language resources in NLP.
  5. Become able to apply practical knowledge of NLP to complex tasks in data analytics.
Assessments
  • Written Assessment (50%)
  • Continuous Assessment (50%)
Module Director
Lecturers / Tutors
The above information outlines module CT5121: "Advanced Topics in Natural Language Processing" and is valid from 2018 onwards.
Note: Module offerings and details may be subject to change.

Optional EE551: Embedded Image Processing


Semester 2 | Credits: 5

This module covers the concepts and technology that are central to embedded image processing: fundamentals of imaging and sensor characteristics; lens characteristics, lens distortion and compensation; basic image processing operations; image filtering, edge detection and segmentation; image compression; image quality assessment. The course material will be supported by practical examples and laboratories/assignments using Matlab and OpenCV software; students will be expected to have some proficiency in Matlab and C or C++ programming.
(Language of instruction: English)

Learning Outcomes
  1. Describe a digital image in terms of the image parameters, sensor characteristics, lens characteristics and colour space.
  2. Perform matrix and vector calculations, and other low-level operations, for image processing functions
  3. Describe and apply frequency domain filtering techniques to digital images.
  4. Apply segmentation algorithms to digital images.
  5. Develop and apply feature detection algorithms to digital images.
  6. Apply image compression algorithms to raw digital images and evaluate the effects of those algorithms, e.g. in terms of image quality.
  7. Effectively apply OpenCV to implement a range of image processing operations.
Assessments
  • Continuous Assessment (100%)
Module Director
Lecturers / Tutors
The above information outlines module EE551: "Embedded Image Processing" and is valid from 2016 onwards.
Note: Module offerings and details may be subject to change.

Optional CT5113: Web and Network Science


Semester 2 | Credits: 5

This module will provide the student with the skills to extract, clean and analyse data from the Web. The focus will be graph and network analytic approaches to Web-mining. Topics include: graph theory, network modeling, social network analysis, community-finding techniques, models of information diffusion, link prediction, evaluation techniques. There will be practical sessions on using graph-data bases and graph visualisation tools such as Gephi. The student will learn how to apply Web mining techniques to applications such as recommender systems, adaptive personalisation, authority ranking.
(Language of instruction: English)

Learning Outcomes
  1. know how to collect, extract and preprocess data in a variety of web formats and standards
  2. understand the distinctions and techniques used for the 3 different aspect to mining web data: Web usage mining, Web content mining and Web structure mining.
  3. know how to extract and use patterns from hyperlink structure of the Web
  4. know how to extract and interpret patterns from user interaction networks generated by Web social media using techniques from social network analysis
  5. know how to link unstructured Web content to structured Linked Data
  6. understand how to apply Web mining techniques to applications such as recommender systems, adaptive personalisation and opinion mining
  7. understand the fundamentals of modelling and an analysing information in the form of a graph
  8. understand the principles behind graph partitioning, community-finding and modularity
  9. understand the principles and main models in information diffusion and know how to apply these
  10. understand the key ideas and techniques used in link prediction
Assessments
  • Written Assessment (40%)
  • Continuous Assessment (60%)
Module Director
Lecturers / Tutors
Reading List
  1. "Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data" by Bing Liu
    ISBN: 9783642194597.
    Publisher: Springer
The above information outlines module CT5113: "Web and Network Science" and is valid from 2018 onwards.
Note: Module offerings and details may be subject to change.

Optional CT5137: Knowledge Representation & Statistical Relational Learning


Semester 2 | Credits: 5

This module introduces students to formal Knowledge Representation and Statistical Relational Learning. Knowledge representation and reasoning are concerned with the efficient formal representation of information and its utilization for automated problem-solving tasks. Statistical Relational Learning is an area of Artificial Intelligence and Machine Learning concerned with the representation of, and reasoning and learning with, uncertain (probabilistic) and relational domain knowledge (such as graphs, web links or symbolic facts). Planned topics: Foundations of knowledge representation. Propositional and first-order logic. Foundations of reasoning (deductive, inductive, abductive, probabilistic). SAT and SMT. Logic programming. Probabilistic logics and uncertainty reasoning. Parameter and structure learning in statistical-relational settings. Requirements: Existing basic knowledge of Machine Learning
(Language of instruction: English)

Learning Outcomes
  1. Understand the fundamental principles of knowledge representation and reasoning
  2. Understand syntax and semantics of important non-probabilistic and probabilistic logics
  3. Understand fundamental types of and approaches to reasoning
  4. Being able to model simple application domains using logic languages and relational knowledge representation formats
  5. Understand and apply fundamental principles of Machine Learning in statistical-relational settings
Assessments
  • Written Assessment (75%)
  • Continuous Assessment (25%)
Module Director
The above information outlines module CT5137: "Knowledge Representation & Statistical Relational Learning" and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.

Why Choose This Course?

Career Opportunities

AI skills will be required in every industry and could create—according to a recent Forbes report—globally up to 22 million new jobs by 2022. The World Economic Forum estimates that by 2025 machines are expected to perform more current work tasks than humans compared to 71% being performed by humans today, and their recent report concludes that artificial intelligence, robotics and smart automation technology could contribute up to $15 trillion to global GDP by 2030.  

Within the AI space, there is a diversity of jobs requiring various levels of expertise:

  • More foundational jobs include data architects, software engineers and machine and deep learning engineers.
  • Advanced roles include specialist research engineers, including those that specialise in computer vision, language and speech, and AI architects.

Who’s Suited to This Course

Learning Outcomes

 

Work Placement

Study Abroad

Related Student Organisations

Course Fees

Fees: EU

€7,215 p.a. 2019/20

Fees: Tuition

€6,991 p.a. 2019/20

Fees: Student levy

€224 p.a. 2019/20

Fees: Non EU

€17,750 p.a. 2019/20

Find out More

MSc (AI) Programme Administrator,
Discipline of Information Technology,
College of Engineering and Informatics,
NUI Galway.
T: +353 91 493 836
E: MScCS-AI@nuigalway.ie
www.it.nuigalway.ie/

Downloads

  • Postgraduate Taught Prospectus 2020

    Postgraduate Taught Prospectus 2020 PDF (21 MB)

  • AI brochure 2019

    AI brochure 2019 PDF (680kb)