London School

LDS7003M Assignment Brief (Component 2)

Contents
Module Details ........................................................................................................... 1
Assignment Description.............................................................................................. 2
Learning Outcomes .................................................................................................... 3
Advice and Guidance ................................................................................................. 4
How is this assessment marked? ............................................................................... 5
Marking Criteria .......................................................................................................... 6

Module Details

Module code: LDS7003M- Level of Study: 7

Artificial Intelligence
and Machine
Learning
Module Leader(s): Swathi Ganesan Credits: 20

Assessment format: Presentation- Group Method of In class and on

presentation (15 submission: Turnitin within Moodle
minutes) of ML case
with individual
preparation
component
Deadline or 19th Dec 2024, Feedback date and 22 Jan 2025, Turnitin
Assessment Period: 12Noon place: within Moodle
submission
Assessment limits: N/A Component number: 2 of 2
length, load, word count,
etc.
Is this exempt from Yes Component 40%
anonymous marking weighting:
under the policy?

1
 London School

Assignment Description

In this group project, you will use machine learning techniques to tackle a real-world problem. Your
primary objectives are to explore, preprocess, model, and evaluate various machine learning
algorithms, and then critically analyse their performance in the context of your chosen domain. The
goal is to predict a continuous numerical outcome based on multiple input features.

Scenario: Choose a dataset where the task involves predicting a continuous numerical outcome.
Possible scenarios include forecasting house prices, estimating sales revenue, or predicting patient
medical costs.

Dataset Selection: Select one of the following datasets for your analysis:

1. House Prices Dataset: This dataset contains 79 explanatory variables describing various
aspects of residential homes. The challenge is to predict the final price of each home.

2. Bike Rental Dataset: This dataset is aimed at predicting the total number of bike rentals (a
continuous variable) based on weather data, seasonal information, and other factors.

3. Medical Cost Personal Dataset: This dataset involves predicting the future medical costs of
patients based on their personal attributes such as age, BMI, smoking status, and more.

Datasets are available to download from Moodle.

Task:

1. Data Exploration and Preprocessing:

• Conduct exploratory data analysis (EDA) to identify trends, correlations, and outliers.
• Handle missing data and perform any necessary data cleaning.
• Perform feature engineering to create new relevant features.
• Scale and normalize the data as needed.

2. Model Development:

• Implement at least three different algorithms (e.g., Multiple Linear Regression, Support
Vector Regression, etc.,)
• Use k-fold cross-validation to assess the performance of each model.

3. Model Evaluation:

• Compare the models using appropriate regression metrics such as RMSE, MAE, and R-
squared.
• Use Recursive Feature Elimination (RFE) or another feature selection method to identify the
most important features.
• Perform hyperparameter tuning using GridSearch or another hyperparameter tuning method
to optimize the models.

4. Domain and Ethical Implications:

2
 London School

Assignment Description

• Discuss how the models can be applied in the context of your chosen dataset
• Explore potential impacts and applications of your findings in business or research scenarios.
• Analyse ethical considerations related to your dataset and predictive modelling. Address
issues such as data privacy, informed consent, and the potential consequences of model
predictions

Learning Outcomes
You must successfully achieve the following Learning Outcomes to pass this assessment:

PLOs 7.1-7.5, 7.8, 7.9

7.1 Critically apply skills, techniques, and knowledge from a range of data analysis methods and
algorithms for enhancing and solving problems in various domains.
7.2 Develop abstract thinking and design ability to analytically demonstrate concepts relating to data
science.
7.3 Use research-based knowledge for the design of experiments, analysis, and interpretation of data
to provide valid results.
7.4 Critically evaluate and analyse advanced data science topics, and concepts, and implement them
in workplace.
7.5 Identify and implement appropriate programming and software tools to critically analyse big data
applications in workplace.
7.8 Critically analyse the data and apply predictive modelling technique in the field of Machine
Learning and Artificial Intelligence.
7.9 Critique legal, social, and ethical issues within the field of data science and applicable ancillary
sectors, as applied to contemporary research and industrial practice.

Advice and Guidance

Assessment Guidance

Presentation Guidelines
• Presentation should include an introduction to the problem, a summary of your data
exploration and preprocessing steps, details of your model development and evaluation, and
a discussion of the domain and ethical implications.
• Use charts, graphs, and tables to illustrate key points and findings. Ensure that visuals are
clear and effectively communicate your results.
• Provide a thoughtful critique of your models' performance, including any limitations and
areas for improvement.
• Prepare to answer questions and engage in discussion about your methods and results.

3
 London School

Advice and Guidance

Remember, this is a group presentation and each group member should contribute to researching
and preparing different parts of the assignment to ensure a well-rounded and comprehensive
exploration of the findings.

Deliverables
• Presentation Slides: Submit a deck that includes all necessary components of your project
in Moodle.
• Code and Documentation: Provide a link to your code repository and include
documentation on how to run your analysis.

General Guidance
General considerations
Please be aware that each step should be fully described in your assessment. Collaborate with your
group members effectively but ensure your individual understanding of the assignment and present
the individual task you are working on. Students should work together to ensure a cohesive and well-
coordinated presentation. Each individual’s contribution should complement the others, creating a
comprehensive and coherent narrative.

Additional Information
The work you present should be your own work, and not just copied from others. You can quote
from others, but you must say who the author is and use quotation marks or paraphrase. If you do
not do so, we will investigate your work for academic misconduct. This is particularly likely if your
Turnitin similarity score is above 25% and/or individual matches are above 6%.

If you require support with your study skills, please visit https://www.yorksj.ac.uk/students/study-skills/

Assessment Regulations
Please refer to the York St John University Code of Practice for Assessment and Academic Related
Matters 2024-25.

We ask that you pay particular attention to the academic misconduct policy. Penalties will be applied
where a student is found guilty of academic and/or ethical misconduct, including termination of
programme (Policy Link).

You are required to keep to the word limit set for an assessment and to note that you may be subject
to penalty if you exceed that limit. You are required to provide an accurate word count on the cover
sheet for each piece of work you submit (Policy Link).

For late or non-submission of work by the published deadline or an approved extended deadline, a
mark of 0NS will be recorded. Where a re-assessment opportunity exists, a student will normally be
permitted only one attempt to be re-assessed for a capped mark (Policy Link).

An extension to the published deadline may be granted to an individual student if they meet the
eligibility criteria of the (Policy Link).

4
 London School

How is this assessment marked?

Your work will be marked according to the assessment instructions provided within this document
and the selected Learning Outcomes’ (LOs) (see above).

Furthermore, this assessment is marked using the assessment marking criteria or a similar rubric
that aligns with the University’s Generic Assessment Descriptors (see below). 1 This is to ensure all
assessment decisions are comparable regardless of the discipline or mode of assessment.

Please note that you must meet the required baseline standards (50 – 59%) which will include the
LOs and minimum expectations of the assessment. Further still, you must ensure you meet the
requirements of each grade boundary to progress to the next, i.e., you should demonstrate your
learning through the standards of the Pass, Merit and Distinction to reach a Distinction (70 – 84%).
These standards are designed to scaffold and build your learning to achieve your fullest potential in
each criterion being assessed.

1 A rubric is a type of scoring guide that markers use to set out specific components and expectations for an assignment for their students.
It is then used to guide the marking they undertake.

5
 London School

Marking Criteria
Pass Grade Bands (100 – 50) (Learning Outcomes must be met)
Fail Grade Bands (49 – 0) (Learning Outcomes are not met)

Mark
Assessment Criteria Description
(100%)

Demonstrate a clear understanding of the problem domain and the relevance of predicting a continuous numerical
Problem
outcome based on input features. Articulate the significance and objectives of the analysis in the context of the chosen 10%
Comprehension
dataset.

Knowledge and Demonstrate understanding of data exploration techniques, including EDA, handling missing data, data cleaning, and
15%
Execution feature engineering. Appropriateness and effectiveness of preprocessing steps applied.

Implementation and Apply at least three different algorithms. Show an understanding of the models’ workings and provide a rationale for
25%
Variety their selection. Demonstrate proper use of k-fold cross-validation for performance assessment.

Compare models using appropriate regression metrics (RMSE, MAE, R-squared, etc). Use feature selection methods
Evaluation and
and hyperparameter tuning to optimise model performance. Critical analysis of model results and selection of the best- 25%
Optimization
performing model.

Discuss how the models can be applied in the context of the chosen dataset. Explore potential impacts and
applications in business or research scenarios and demonstrate a clear understanding of the practical implications.
Application and
10%
Ethical Impact
Analyse ethical considerations related to data privacy, security, and the responsible use of predictive models. Address
any potential biases and unintended consequences. Provide strategies to mitigate these issues.

Quality of Assess readability, grammar, structure, and completeness. Evaluate the effectiveness of communication, collaboration
15%
Presentation with the team, use of visuals, and overall presentation skills.

Overall Total 100%

6
 London School

Level 7 GAD Descriptor for Assessment Matrix

7
 London School

Distinction (70 – Distinction (85 – Borderline Fail

Assessment Criteria Pass (50 – 59) Merit (60 – 69) Fail (30 - 44) Fail (0 - 29)
84) 100) (45 - 49)
Demonstrates a basic Shows a good Provides a thorough Demonstrates an
Limited Fails to adequately
understanding of the understanding, understanding, exceptional grasp of No clear
understanding of comprehend the
Problem problem domain and clearly articulating offering insightful the problem, understanding of
Thinking Skills & the problem problem domain or
Comprehension the significance of the problem's connections presenting a detailed the problem
Research Skills domain with basic articulate the
(10%) predicting the outcome relevance and the between the analysis of its domain or
articulation of significance of the
based on input objectives of the problem and the relevance and objectives.
relevance. analysis.
features. analysis. dataset used. objectives.
Applies some data
Applies basic data Demonstrates solid Shows advanced Excels in applying
exploration Limited or Neglects data
exploration techniques, understanding and understanding and data exploration
techniques but inappropriate use exploration and
Knowledge and Thinking Skills & including EDA and effective execution of application of data techniques and
lacks depth in data of data exploration preprocessing,
Execution (15%) Research Skills basic data cleaning data exploration and exploration, feature preprocessing with a
cleaning and and preprocessing leading to poor
and feature preprocessing engineering, and deep understanding
feature techniques. execution.
engineering. techniques. data preprocessing. of their impact.
engineering.
Effectively Demonstrates a
Expertly applies a Inadequate Fails to implement
Implements relevant implements and high level of Applies algorithms
variety of algorithms, implementation of the algorithms
Implementation Practical Skills & algorithms with basic justifies the choice of understanding in but with a limited
providing a algorithms, with correctly, showing
and Variety Professional understanding and the algorithms used, applying multiple understanding of
sophisticated little understanding no understanding
(25%) Learning Skills appropriate use of k- using k-fold cross- algorithms and model selection
rationale and precise of model selection of model selection
fold cross-validation. validation cross-validation and validation.
cross-validation. and validation. or validation.
appropriately. techniques.
Provides a detailed
Effectively compares
evaluation of No meaningful
models, Excels in evaluating Fails to properly
Compares models models with in- comparison or
demonstrates a solid and optimizing Basic comparison compare or
Evaluation and Practical Skills & using basic regression depth analysis, optimization of
understanding of models with a critical of models with optimize models,
Optimization Professional metrics and attempts effectively models; lacks
optimization through and innovative limited optimization with inadequate
(25%) Learning Skills feature selection and optimizing understanding of
feature selection and approach to feature techniques. use of evaluation
hyperparameter tuning. performance evaluation
hyperparameter selection and tuning. metrics.
through advanced metrics.
tuning.
methods.
Discusses basic Provides a thoughtful Demonstrates a Provides a Inadequate Fails to discuss
Limited discussion
applications of the discussion of the strong comprehensive and discussion of the practical
on applications and
Application and models in a real-world models' applications understanding of innovative models' applications or
Thinking Skills & ethical impacts,
Ethical Impact context, with limited and ethical impacts, how the models can discussion on the applications and ethical issues,
Research Skills lacking depth and
(10%) exploration of ethical with relevant be applied with practical and ethical ethical issues, with showing no
actionable
impacts. Identifies strategies to mitigate insightful analysis of implications, little consideration understanding of
strategies.
basic ethical potential issues. ethical presenting advanced of impact. their relevance.

8
 London School
Distinction (70 – Distinction (85 – Borderline Fail
Assessment Criteria Pass (50 – 59) Merit (60 – 69) Fail (30 - 44) Fail (0 - 29)
84) 100) (45 - 49)

considerations with considerations and strategies to mitigate

some strategies for implications. risks.
mitigation.
The presentation is The presentation is
The presentation is The presentation is The presentation
clear with basic exemplary, with The presentation is The presentation
effective, with good highly professional, is unclear, poorly
Quality of readability, structure, outstanding clarity, comprehensible lacks clarity and
Communication, readability, well- with excellent structured, and
Presentation and appropriate use of structure, and but may lack focus structure, with
Collaboration structured content, readability, logical fails to
(15%) visuals. Shows minimal innovative use of or technical limited use of
and effective use of structure, and communicate
creativity in developing visuals to enhance proficiency. visuals.
visuals. impactful visuals. effectively.
the presentation. understanding.