Machine Learning


Owner: Dr Dorian Florescu
Number of students: 2
Formative Deadline: Friday W26

Learning Outcomes

Upon successful completion of this skill, students will be able to:

  • Demonstrate familiarity with core concepts in machine learning.
  • Apply appropriate machine learning methods—classification, regression, or unsupervised learning—to analyse new datasets beyond those covered in the taught material.
  • Synthesise multiple analytical approaches, including classification, regression, and unsupervised learning, to comprehensively evaluate new datasets beyond those covered in the taught material

Knowledge Requirements

You will need to have competed all four courses at the link below, namely:

  1. Classification Methods with Machine Learning
  2. Regression Methods with Machine Learning
  3. Cluster Analysis with Machine Learning
  4. Dimensionality Reduction Techniques

https://matlabacademy.mathworks.com/details/machine-learning-techniques-in-matlab/lpmlmlt

Please upload the pdfs of all four Course Completion Certificates generated after each course is fully completed.

Application Requirements

To claim this skill, you will use a single Machine Learning (ML) model to perform a task on a new chosen dataset, beyond those covered in the taught material, using Python or MATLAB. The task should be in one of the following categories: classification, regression or unsupervised learning.

Before attempting to claim this skill, you need to have competed all four courses at https://matlabacademy.mathworks.com/details/machine-learning-techniques-in-matlab/lpmlmlt, and uploaded the corresponding Matlab certificates. You should show:

  1. The pdfs of all four Course Completion Certificates generated after each course is fully completed.

  2. A brief description of the aims of the study, and the choice of the particular ML model (50-100 words).
  3. A brief explanation (50-100 words) of what makes the chosen dataset suitable for the category of task selected (classification/regression/unsupervised learning).
  4. Screenshots of the training process clearly marked with the student’s username, accompanied by a brief description of the results, including relevant error metrics (50-100 words).
  5. The equation representing the weight/parameter update rule for the model of choice, including values for all constants. If not applicable, a brief (50 words) explanation of why it is not.
  6. A description of the hyperparameters that were optimised, the corresponding ranges, and brief interpretation of why the final hyperparameters might be a better choice (50-100 words).

  7. A concise discussion (50-100 words) will also be included commenting on further analyses/improvements could be made to the chosen approach. An additional requirement is that the skill is claimed using the Mahara template entitled “Y2 Machine Learning Application Template”.

Synthesis Requirements

To claim this skill, you will use a single Machine Learning (ML) model to perform a thorough analysis on a new chosen dataset, beyond those covered in the taught material, using Python or MATLAB. The analysis should cover all the following categories: classification, regression or unsupervised learning.

Before attempting this, make sure that you have fulfilled the skill requirements at application level.

You should demonstrate the application-level requirements 2-7 for all the three analysis categories (classification/regression/unsupervised learning).

An additional requirement is that the skill is claimed using the Mahara template entitled “Y2 Machine Learning Synthesis Template”.

Knowledge Opportunities

Lab W22

Application Opportunities

Lab W25


Labs Demonstrating This Skill

Labs Contributing Towards This Skill