Where to Walk

An interactive dashboard telling the story of Melbourne, Australia's foot traffic.

Predictions are made using 49 negative binomial models (one for each sensor that had adequate data).

The models were selected for explainability and goodness of fit.

Included visualisations:

• A line plot showing the pedestrian count at different times of day across different sensors
• A boxplot comparing the variation in pedestrians for each day of the week
• A map highlighting the areas of Melbourne with the most foot traffic at any given time and day

Made using:

• R
• Shiny
• Leaflet
• ggplot
• plotly

Data sourced from Melbourne City Open Data:

Sensor Location Data
Pedestrian Count Data