School of Physics, Engineering and Computer Science
Weighting %: 30% Submission deadline See Canvas
(for students):
Authorship: Individual Target date for
returning marked
coursework:
Tutor setting the work: Andy Tai Hock Lin Number of hours you 15
are expected to work
on this assignment:
This Assignment assesses the following module Learning Outcomes (from Definitive Module
Document):
1. Demonstrate knowledge and understanding of the fundamental mathematical ideas behind data
science.
2. Demonstrate knowledge and understanding of relevant computational algorithms and the
fundamentals of probability, information and statistical methods.
3. Demonstrate knowledge and understanding of producing appropriate algorithmic definitions to
provide useful for data analytics.
Assignment Tasks:
Coding assignment and written report
This assignment will test your ability to analyse a dataset and apply appropriate computational algorithms
to elicit insights and knowledge.
You will be working on the insurance cost Multiple Linear Regression dataset available at:
[Link]
This dataset contains medical insurance cost information for 1338 individuals. It includes demographic
and health-related variables such as age, sex, BMI, number of children, smoking status, and residential
region in the US. The target variable is charges, which represents the medical insurance cost billed to the
individual.
Assignment Requirements
1. Dataset Understanding and framing the problem
• Download the dataset from the provided link.
• Provide a brief description, including the source, number of records, and key variables.
• Frame the problem, and provide rationale for regression (and/or other modelling of choice)
2. Data preprocessing and feature engineering
• Perform relevant data cleansing (handling missing values, correcting data types).
• Carry out feature engineering where appropriate (e.g., creating derived variables such as total
study time or binary indicators).
3. Exploratory Data Analysis (EDA)
• Conduct statistical summaries and visualisations to identify trends and correlations (e.g.,
relationship between study hours and final score).
• Discuss notable insights (e.g., which variables seem to have the strongest impact on
performance).
4. Algorithm Implementation (Model Development)
• Implement at least two computational algorithms from the following (or others, with
justification):
o Multiple Linear Regression (mandatory)
o Decision Tree Regression or Classification
o Random Forest Regression or Classification
o Clustering models (if you choose to segment students based on characteristics)
Page 1 of 3
School of Physics, Engineering and Computer Science
o Principal Component Analysis (PCA)
• Justify your choice of algorithms.
5. Evaluation of Results (Evaluation and Interpretation)
• Use appropriate evaluation metrics such as R², MAE, RMSE for regression tasks, and
accuracy/F1-score for classification tasks.
• Compare algorithm performances and interpret results.
6. Presentation of Findings (Insights and Reporting)
• Summarise your findings in a well-structured report including:
o Visual aids (graphs, tables, and charts)
o Actionable recommendations (e.g., strategies to improve student performance based
on model findings).
7. Code Quality and Documentation
• Ensure that your code is well-commented, readable and modular
Submission Requirements:
• Code submission - You should submit your completed code as a Python script (.py) or notebook
as a .ipynb, Ensure that the code is well-documented with comments explaining key sections.
• Report Submission - You should also submit your report as a .pdf or .docx. The report should
include an introduction, methodology, EDA, algorithm implementation, results, discussion, and
conclusion.
• Dataset – Include the downloaded dataset or link to the Kaggle source.
Marks awarded for:
Type of Feedback to be given for this assignment:
Written feedback on the code and report.
Additional information:
Page 2 of 3
School of Physics, Engineering and Computer Science
• Regulations governing assessment offences including Plagiarism and Collusion are available from
[Link]
[Link] (UPR AS14).
• Guidance on avoiding plagiarism can be found here: [Link]
(see the Referencing section)
• Adhere to UH’s policy on use of GenAI:
o “Unauthorised use of artificially generated material (AI) in researching or presenting material for
an assessment is an academic misconduct offence if you use AI tools in producing your
assessment, unless the use of AI tools is expressly permitted. However, even if expressly
permitted, where you do not declare that you have used an artificial intelligence tool(s) in the
production of your assessment, or you are dishonest about the extent to which such tools have
been used, you will have committed academic misconduct.”
• For postgraduate modules:
o a score of 50% or above represents a pass mark.
o late submission of any item of coursework for each day or part thereof (or for hard copy
submission only, working day or part thereof) for up to five days after the published
deadline, coursework relating to modules at Level 7 submitted late (including deferred
coursework, but with the exception of referred coursework), will have the numeric grade
reduced by 10 grade points until or unless the numeric grade reaches or is 50. Where the
numeric grade awarded for the assessment is less than 50, no lateness penalty will be
applied.
Page 3 of 3