Prepare
Use documented median/mode imputation.
↗← ALL WORKLook beyond
ACADEMIC PROJECT / PYTHON + PANDAS / GITHUB / 05
Look beyond
the asking price.
An exploratory view of Melbourne property data, 2016–2018.
EXPLORE THE STORY ↓≈0.47reported rooms–price correlation
A descriptive relationship, not a causal effect.
01
THE BUSINESS PROBLEM
Which property features move with price?
Explore relationships while keeping missing values, outliers and genuine luxury properties in context.
≈0.47reported Pearson correlation
Strongest reported price correlation among the examined variables. Association, not causation or a predictive accuracy score.
02From question
WHAT I DID
From question
to useful work.
Inspect
Apply percentile-based capping and review the trade-offs.
↗Explain
Visualise Pearson relationships with Python and communicate the limits.
↗03A documented pattern.
THE RESULTS
A documented pattern.
A careful interpretation.
Rooms had the strongest reported price correlation among the variables examined. The work makes data-cleaning choices part of the story.
2016–18
Data period explored.
≈0.47
Reported rooms–price correlation.
Python
Reproducible exploratory workflow.
THE BUSINESS TAKEAWAY
Treat correlation as a starting point for further modelling, not a guaranteed price effect.
↗THE EVIDENCE / GO DEEPER
LET’S DISCUSS YOUR PROJECT ↗Methods, context & full project notes ＋
The business question
Which property characteristics are associated with price, and how do cleaning choices influence the interpretation?
The work
The repository describes median/mode imputation, percentile-based outlier capping, Pearson correlations, and Matplotlib/Seaborn visualisation. It also recognises that high-value observations can represent genuine luxury properties rather than errors.
The finding
The README reports that rooms had the strongest price correlation among the variables examined, at approximately 0.47, with a slight decline in average price across the period. Correlation is descriptive and does not establish a causal price effect.
What the project demonstrates
A documented exploratory workflow, attention to data-quality trade-offs, and the ability to communicate patterns before moving to predictive modelling.
Source: public GitHub project README and linked academic reports.
Customer retention ↗
YOUR NEXT MOVE