Machine Learning

Learning Objectives

Learn about the key paradigms and algorithms in machine learning.
Get an understanding of data analytics based on machine learning and using modern programming tools, such as Python or R.
Experience how machine learning and data analytics can be used in real-world applications.
Acquire the ability to gather and synthesise information from multiple sources to aid in the systematic analysis of complex problems using machine learning tools and algorithms.
Articulate the legal, social, ethical, and professional issues faced by machine learning professionals.
Understand the applicability and challenges associated with different datasets for the use of machine learning algorithms.
Apply and critically appraise machine learning techniques to real-world problems, particularly where technical risk and uncertainty is involved.
Systematically develop and implement the skills required to be an effective member of a development team in a virtual professional environment, adopting real-life perspectives on team roles and organisation.

Group Assignment

Unit 8, Business analytical questions, Group Briefing and Notebook code.

Individual Presentation

Unit 11, Neural Network (CIFAR 10), Presentation and Code.

Unit 1, Introduction to Machine Learning (ML)

In this unit, I learned about the past, present, and future of machine learning, the challenges and opportunities of using algorithms, and the interplay among big data analytics, machine learning, and artificial intelligence. By the end of the week, I had a better understanding of the role of machine learning in the future industry, identified the skill sets required to become proficient in machine learning, and learned about the pitfalls of machine learning and ways to address them.

Contributed my ideas to the collaborative discussion forum on our learning platform. I discussed the ‘Fourth Industrial Revolution’ and its likely impact on Local Government, with AI likely to help Adult Social Care services which cost £20.5 billion annually for the UK (Parliament 2024).

Link to my writing is here.

Unit 2, Exploratory Data Analysis (EDA)

I learned about the steps involved with Exploratory Data Analysis (EDA), which included understanding the dataset through feature exploration and spotting anomalies within the data. I focused on visual analysis techniques to gain deeper insights into the data. By the end of the week, I was able to undertake basic EDA, understand the dataset thoroughly, and prepare it effectively for machine learning. This comprehensive learning experience equipped me with the skills needed to handle datasets confidently and ensure their readiness for further analysis and modelling.

Unit 3, Correlation and Regression

In this unit, we looked at the core aspects of correlation and regression, including techniques to measure them and their application through examples. I gained a theoretical understanding of correlation and regression and learned how to compute these statistical techniques. By the end of the week, I was able to apply correlation and regression in real-world scenarios, using these methods to analyse and interpret data effectively.

Unit 4, Linear Regression with Scikit-Learn

For unit 4, I learned how to use Scikit-Learn to model a linear relationship and develop a multivariate linear regression model. I also focused on evaluating the effectiveness of my models. By the end of the week, I was able to undertake regression modelling with complex datasets, evaluate the results to optimize the models, and gain a deeper understanding of the Scikit-Learn library in Python.

Unit 5 and 6, Clustering & Clustering with Python

Studied about the basic concept of clustering, various techniques of distance measurements, and specific methods like K-means and agglomerative clustering. I also delved into how to effectively evaluate clusters once they're formed. By the end of the week, I had a solid understanding of the logic behind clustering, could identify the necessary skills to evaluate cluster analysis results, and gained insight into potential pitfalls associated with clustering techniques. This learning experience equipped me with the knowledge and tools to apply clustering methods effectively in practical data analysis scenarios.

Unit 7, Introduction to Artificial Neural Networks (ANNs)

Explored the analogy between biological neurons and artificial neural networks (ANN), gaining a detailed understanding of the algorithms that underpin ANN and the various functions used within them. By the end of the week, I had acquired a critical and comprehensive understanding of ANN, allowing me to design and develop simple ANN models. Additionally, I developed the ability to critique and contextualise emerging research in the field of ANN, enhancing my capacity to apply this knowledge to practical and theoretical scenarios.

Unit 8, Training an Artificial Neural Network

For this unit, I focused on understanding how artificial neural networks (ANN) learn from errors, utilising backpropagation to adjust neuron connection weights effectively. I explored various real-life applications of ANN, illustrating their practical relevance. By the end of the week, I gained insights into the error handling mechanisms of ANN, enabling me to design and develop more sophisticated ANN models. Furthermore, I developed skills to critique and contextualize emerging research within the field of ANN, enhancing my ability to apply this knowledge in both theoretical and practical contexts.

Collaborative Discussion 2, e-Portfolio Activity: Gradient Cost Function

Link to my writing is here.

Unit 9, Introduction to Convolutional Neural Networks (CNNs)

This week, I delved into the detailed algorithm underlying Convolutional Neural Networks (CNNs), explored specific Python libraries tailored for CNNs with practical examples, and examined real-life applications of CNNs. By the end of the unit, I grasped the application and significance of computer vision, equipped to perform fundamental computer vision tasks confidently. This learning experience provided me with the foundational knowledge and practical skills necessary to engage with CNNs effectively in various applications and scenarios.

Unit 10, CNN Interactive Learning

Centred on exploring the structure and gaining an in-depth understanding of Convolutional Neural Networks (CNNs). I delved into the intricacies of CNNs, learning about their architecture and how they function in computer vision tasks. By the end of the module, I achieved a solid grasp of the application and significance of computer vision, and I developed the ability to perform basic tasks related to computer vision confidently. This learning experience equipped me with the foundational knowledge needed to engage effectively with CNNs in practical applications and scenarios involving visual data processing.

Unit 11, CNN Interactive Learning

Studied the workflow of model selection and evaluation, learning about the approach and techniques used to select appropriate models for specific tasks. I also explored methods to effectively evaluate models and strategies to improve their performance systematically. By the end of the module, I gained a clear understanding of the critical importance of model selection and optimisation in data science and machine learning projects. This learning experience equipped me with practical insights and tools to make informed decisions when selecting and refining models for various applications.

Summative Assessment: Individual Presentation

Link to Presentation and Code.

Unit 12, Industry 4.0 and Machine Learning

Explored deeper into the ‘fourth industrial revolution’ and the spread of machine-derived data, understanding how machine learning algorithms are deployed within this evolving environment. I also delved into the concept of digital twins and their connection to data science, examining their role in virtual simulations and predictive analytics. By the end of the unit, I developed a comprehensive understanding of the challenges and opportunities presented by the ‘fourth industrial revolution’ and gained insights into the profound impact of machine learning on future societies. This learning experience equipped me to appreciate the transformative potential of advanced technologies in shaping industries and societies globally.

References

UK Parliament (2024) ‘Funding for adult social care in England’ Available at: https://commonslibrary.parliament.uk/re