Unit 26: Machine Learning

Unit code                            J/615/1662

Unit level                            QCF Level 5/ NFQ Level 6/7

Credit value                       15


Machine learning is the science of getting computers with the ability to learn from data or experience to solve a given problem without being explic itly programmed.

It has been around for many  years, however it has become one of the hottest fields of study in the computing sector. Machine learning is in use in several areas such as predictive modelling, speech recognition, object recognition, computer vision, anomaly detection, medic al diagnosis and prognosis, robot control, time series forecasting and much more.

This unit will introduce the basic theory of machine learning,  the most  efficient machine learning algorithms and practical imple mentation of these algorithms. Students will gain hands-on experience in getting these algorithms to solve real- world problems.

Topics included in this unit are: the foundations of mac hine learning, types of learning problems (classification, regression, clustering etc.), taxonomy of mac hine learning algorithms (supervised learning, unsupervised learning, reinforcement learning), mac hine learning algorithms (Decision Tree, Naïve Bayes, k-Nearest Neighbour,  Support Vector Machine etc.).

On successful completion of this unit students will be able to understand the concept of machine learning, mac hine learning algorithms,  gain hands-on experience in imple menting algorithms using a programming language such as C/C++, C#, Java, Python, R, or a mac hine learning tool such as Weka, KNIME, MS AzureML etc.

As a result students will develop skills such as communic ation literacy, critical thinking, analysis, reasoning and interpretation, which are crucial for gaining employment and developing academic competence.


Learning  Outcomes

By  the end of this unit students will be able to:

LO1.     Analyse the theoretical foundation of mac hine learning to determine how an intelligent mac hine works.

LO2.     Investigate the most popular and efficient mac hine  learning algorithms used in industry.

LO3.     Develop a mac hine learning applic ation using an appropriate programming language or mac hine learning  tool for solving a real-world problem.

LO4.     Evaluate the outcome or the result of the application to determine the effectiveness of the learning algorithm used in the application.