THE PROJECT NOTEBOOK · 09 ENTRIES
A question.
A dataset. A discovery.
Explorations in data cleaning, analysis, and visual storytelling.
TTC ridership: a recovery that hasn't caught up
Two decades of Toronto transit data: the 2016 peak, the pandemic collapse, and why measuring recovery starts with checking what a month means.
What shapes a student’s performance?
Mapping daily study habits, sleep, screen time, and wellbeing against exam outcomes across 1,000 simulated student records to find what actually moves the needle.
When the data doesn’t tell a story
A predict-and-reveal exploration of 8,000 synthetic user records. Make a prediction, compare churn rates with counts, and discover why the denominator changes the story.
120 years of going for gold
Exploring participation growth, gender representation, and medal concentration across 70,000 Olympic athlete-event records from 1896 to 2016.
The patterns behind the price tag
Analyzing pricing strategy, discount tiers, and customer review behavior across 1,465 Amazon product listings—and why steep discounts don't buy higher ratings.
Finding signal in diagnostic data
Evaluating 30 cell-nuclei measurements across 569 diagnostic samples from the Wisconsin Breast Cancer dataset to isolate the strongest predictors of malignancy.
A cleaner cup of data
Reconstructing missing prices, quantities, and totals across 10,000 noisy cafe transactions using deterministic column relationships instead of blind row deletion.
Behind the job description
Turning 672 raw Glassdoor job postings—with sentinel '-1' values, composite salary strings, and unstructured descriptions—into a structured, query-ready dataset.
Making messy transactions make sense
Standardizing schemas, resolving inconsistent category labels, and recovering missing values across 12,575 multi-location retail transactions.