55-602446-BF Machine Learning Algorithms and Heruistics Assessment Brief 2025

55-602446-BF Module Learning Outcomes

• Understand and implement commonly used machine learning algorithms.
• Apply machine learning techniques to develop applications which answer complex questions within a variety of contexts including business, industry and the social media domain.
• Analyse and evaluate techniques for prediction, ranking and recommending using technical and contextual reasoning.

55-602446-BF Assessment Brief

Introduction

This assignment aims to assess your practical understanding of machine learning algorithms by implementing systems (using supervised learning, unsupervised learning and image processing), which can answer complex questions and make intelligent decisions, in the context of the dataset provided and a copyright free image dataset of your choice from Kaggle (Please ensure that you do not use a dataset used in lectures or tutorials).

This is an individual project. You are expected to demonstrate your end-to-end machine learning system design and development skills, including data exploration, feature analysis, machine learning, and performance evaluation.
You need to prepare the following content for the assessment:

1. A 12 +10% minutes video presentation demonstrating your machine learning systems. Ensure that you consider both supervised and unsupervised learning.
   Supervised Learning - Use the complete dataset provided.
   Unsupervised Learning - Delete the label column, to ensure that the data is not labelled. Image Processing - Use a copyright free image dataset of your choice from Kaggle.
2. Supporting documentation for the video including presentation slides, code, and relevant screenshots.

Individual Project

Your machine learning project should include the following components:

For the Dataset Provided:

A. Data Exploration

You need to understand your data first. Your data may require cleaning and pre-processing steps. It will define which algorithms you can choose, and it will ultimately define the performance of your supervised and unsupervised learning models. During the project design and development, you should go through the following steps and try to get as many insights as possible about the data, and for supervised learning its relation to the target variable:

• Analyse the data distribution.

For supervised learning analyse how the target variable is influenced by the features.
Analyse the difficulty of your prediction task.
Please raise any ethical concerns with the tutor.

B. Feature analysis

Define a feature space based on the data samples. Make use of data visualisation tools to find correlated dimensions. You can also measure and analyse features and target dependencies for regression or classification tasks.

C. Machine Learning

Design and develop a suitable learning model using machine learning tools such as Scikit-Learn. Depending on the algorithms (select at least two for supervised learning and two for unsupervised learning), you may need to pre-process the data such as "scaling" and "de-noising". You can use cross-validation approaches to estimate the performance. You also need to keep the balance between under-fitting and over-fitting through adjusting the hyperparameters of the chosen machine learning algorithms.

D. Evaluation

The focus will be the accuracy performance of the developed system. You need to fine-tune your models to improve system accuracy performance. You are expected to compare at least two different algorithms, for each type of learning, in the project. During the system evaluation, you need to analyse and discuss the performance by using at least 2 different scientific evaluation approaches such as ROC curves and confusion matrices.

For the copyright image dataset of your choice from Kaggle (Please get image dataset approved from your tutor before working with it):

A. Data Exploration
Pre-process the dataset selected.
B. Machine Learning
Train a model for classification.
C. Evaluate

Measure accuracy and any other metrics of your choice.

Use the Python programming language and its machine learning packages, such as "Scikit-learn" for your project development.

Your video and supporting documentation should be submitted electronically, through the module's Blackboard site as a single ZIP file, before the deadline (29th April 2025, at 15:00).

55-602446-BF Assessment Criteria:

55-602446-BF University Grade Descriptor