Artificial Intelligence (AI) and Machine Learning (ML) are fast-moving technologies that impacts both our individual lives and society as a whole. The purpose of this module is to give the student a solid grounding of the fundamentals of these technologies, from a human-centric perspective. Core topics in these areas will be enforced with practical examples and the implementation of relevant algorithms in real world problems. Students will develop an understanding of problem solving, knowledge representation, reasoning and learning methods of AI and ML. Throughout the module, practical examples of the use of AI and ML in applications will be demonstrated, and students will consider these from the perspective of the ethical issues involved.
Week 1 : Artificial Intelligence and Machine Learning Core Concepts from an Ethical view
Stages of AI, development of AI and ML, key principles of AI and ML.
Key characteristics of AI, ML and Neural Networks
Types of Learning in ML
Adoption of IA and ML
Ethical considerations in the use of AI and ML techniques
Week 2 - 4: Data Pre-processing and Exploration with Professional Principles
Data collection, data cleaning, and preparation for a safe and transparent analysis
Anonymising datasets
Feature Engineering
Collection methods, identification, and errors removal
Ethical issues in dataset creation and processing
Week 5 - 8: IA and ML techniques with principles of Integrity, Confidentiality and Respect
Parametric Machine Learning Algorithms vs Non-Parametric Machine Learning Algorithms
Theory of generalisation.
Liner classification and regression, logistic regression, gradient descent and future space transformations.
Overfitting and the benefit of regularisation.
Optimal Margin Classifiers and Support Vector Machines.
Week 9 - 10: Model Evaluation and Performance Techniques
Testing Options: Splitting data, Cross-Validation, k-fold cross-validation, Holdout method, confusion matrix, error rate, sampling with replacement, accuracy.
Sensitivity -recall and specificity, precision, f-measure, rood mean square, ROC curve.
Comparing models
Week |1 -12: An Ethical Natural Language Processing
Ethical challenges and social issues in Natural Language Processing
Evaluating Language Models, parsing for NLP
Hidden markov models, part-of-speech tagging
Machine Learning techniques for NLP such naïve bayes, SVM and neural networks - word embedings
Lectures,labs and independent study
| Module Content & Assessment | |
|---|---|
| Assessment Breakdown | % |
| Other Assessment(s) | 100 |