The Australian real estate market is one of the most dynamic and competitive in the world, offering a wide range of properties to both buyers and sellers. For homeowners looking to sell, setting the right price is a critical, and often emotional, decision. After all, property transactions are among the most significant financial events in a person’s life.
Sellers typically set a listing price based on what they believe their home is worth and what the market might bear. But things don’t always go as planned. Some properties attract intense buyer interest and sell for more than the asking price. Others fall short, forcing the seller to accept less than they’d hoped.
If sellers had a way to estimate in advance whether their listed price is likely to be exceeded or undercut,
they could make more informed pricing decisions, better manage expectations, and potentially
maximize their return.
In this assignment, your task is to build a binary classification model that predicts whether a property
will be sold at a higher or lower price than the advertised price set by the seller.
Target Variable
The target variable price_outcome indicates whether a property was sold at a higher, equal or lower
price compared to the listing price.
The values in the price_outcome column are:
• Higher: Sold price is greater than the listed price
• Equal: Sold price is the same as the listed price
• Lower: Sold price is equal to or less than the listed price
This is a binary classification problem; therefore, you should not include any data where the target
value is ‘Equal’. Your model should learn to predict this outcome using the available features of each
properties outlined below.
Dataset
You are provided with a dataset of 6,957 recently sold properties, between February 2022 and February
2023. The predictor variables are:
1. property_address: the address of the property
2. property_suburb : The suburb the property resides in
3. property_state: The state in which the property resides in
4. listing_description: The description of the house provided on the listing
2025 S1 – Assignment
3
5. listed_date: The date the property was listed for sale
6. listed_price: The price the property was listed for
7. days_on_market: The number of days the property was on the market
8. number_of_beds: The number of bedrooms on the property
9. number_of_baths: The number of bathrooms on the property
10. number_of_parks: The number of parking spots on the property
11. property_size: The size of the property in square meters
12. property_classification: The type of property (House/Unit/Land)
13. property_sub_classification: The sub-type of the property
14. suburb_days_on_market: The average days in market that a property is on sale for in a suburb
15. suburb_median_price: The average median property price in a suburb
Deliverables
You must submit the following:
1. A written report (via TurnItIn).
2. A Jupyter Notebook (via the Assignment Submission link).
Your report may be structured as:
• Four main sections: a) Introduction, b) Model Building, c) Model Evaluation, d) Findings &
Conclusion, or
• Three main sections: 1) Introduction, 2) Model Building & Evaluation, 3) Findings &
Conclusion
Both structures are acceptable.
Visuals & Output
✓ You may include up to 8 charts or tables in your report.
✓ All visuals must be supported by the analysis in your Jupyter Notebook.
✓ Your notebook must run without errors — only analysis up to the last successfully run cell will
be marked.
✓ Do not edit the original Assignment_Data.xlsx file before importing.
Formatting and professionalism
Maximum 1500 words (+/- 10%) – including title page, charts and tables.
✓ Use formal language and full sentences (no bullet points).
✓ Times New Roman, 12pt font, single-spaced.
No appendices allowed.
✓ Reports can be written in first person if preferred.
Submission
Submit two files with the following naming convention:
StudentID.pdf and StudentID.ipynb
✓ Written report: via TurnItIn (PDF or DOCX format only)
2025 S1 – Assignment
4
✓ Jupyter Notebook: via Assignment Submission link
Example: If your student ID is 12345678, submit:
12345678.pdf
12345678.ipynb
Do not zip your files.
Note on Academic Integrity
This is an individual assignment. You are encouraged to discuss ideas with your peers but must submit
your own work. Suspected plagiarism or collision will be treated in line with university policy