Module Overview

This module introduces machine learning with a practical approach, providing hands-on cases of real-world data analytics.

It aims to give a conceptual understanding of the underlying theories and practices involved in machine learning. The material evolves gradually, starting with underlying mathematical concepts followed by a framework of learning models. Various paradigms and algorithms are covered thereafter, offering you the theory alongside a wealth of practical case implementation with software. The final stage is a systematic approach to designing a complete application system that makes use of more than one component of machine learning.

Module Units

My Overall Reflection

Introduction

This portfolio provides a reflective overview of my academic journey through the Machine Learning module, the third component of my MSc in Artificial Intelligence. Recognising the central role of this module in the broader programme, I was motivated to engage fully and derive the greatest benefit. With prior modules not entirely meeting my expectations in terms of structure, I was particularly keen for this module to offer a more effective learning experience and to further develop my knowledge and technical skills in the realms of machine learning and deep learning.

General Overview

The initial half of the module revisited foundational machine learning techniques, including regression, clustering, and exploratory data analysis (EDA), which, given my background, did not present significant new challenges. My anticipation grew as the curriculum approached Unit 7, where we explored deep learning, particularly artificial neural networks (ANNs). The final assignment, which involved developing a convolutional neural network (CNN) for image classification, was especially rewarding, as it allowed me to move beyond traditional algorithms and investigate advanced ANN approaches.

Given that the MSc programme is primarily designed for students without a formal background in computer science, I found the module’s overall outcomes to be satisfactory. Nevertheless, I am aware that further independent study will be necessary to deepen my understanding, especially in advanced deep learning. Inspired by what I have learnt, I intend to continue exploring neural networks, expanding from CNNs to other image-based models, such as those for object recognition and image segmentation, as well as delving into deep learning for language processing and sequential data analysis.

Challenges and Overcoming

Throughout this module, I encountered two principal challenges.

The first arose during the team project, which involved constructing a traditional machine learning model using the Airbnb 2019 dataset. The main difficulty lay in coordinating effectively with team members, as we all balanced professional commitments and lived across varying time zones. We addressed this by:

• Maintaining clear and consistent communication among all team members.
• Clarifying roles and allowing members to select responsibilities based on their availability, interests, and learning objectives.

The second challenge was the individual assignment, which required me to design a CNN for image classification, a domain in which I had no prior experience. I overcame this by:

• Beginning my preparation well ahead of the assignment deadline.
• Consulting a range of supplementary resources, including relevant online courses and open lectures.

Successfully completing both assignments was a significant milestone, deepening my theoretical understanding and practical capability in artificial intelligence. This foundation, built on academic study and practical work, positions me well for continued self-directed learning and further development in the field.

The code base and documentation of the team project that I participated in is available here. While the code base and documentation for the individual assignment is available here.

Conclusion

Reflecting on my experience in this module, I am confident that it has equipped me with valuable skills and insights, both technically and in terms of collaborative work. This portfolio showcases the progression of my abilities and my commitment to ongoing professional and academic growth. With a strong foundational knowledge of machine learning and deep learning, I am eager to further my expertise and contribute meaningfully to the field of artificial intelligence.

---

Reference

Rolfe, G., Freshwater, D. and Jasper, M. (2001) Critical Reflection for Nursing and the Helping professions: a user’s Guide. Basingstoke: Palgrave Macmillan.