Skip to content

A repository containning the assets for the data science cycle for predicting fire-risk levels for the greater city of Montreal. Six datasets were leveraged for this project including fire incidents, crimes, property assessments, demographics and weather conditions. Technologies used include Python libraries, Tableau and Excel.

Notifications You must be signed in to change notification settings

ahmedopolis/Montreal-Fire-Prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

High Fire-Risk Areas in Montreal 🔥 🌇 🚒

Python Jupyter Notebook Matplotlib Pandas NumPy scikit-learn SciPy

Summary

The herein repository host the necessary assets for the data exploration, cleaning, integration, feature engineering and modelling for predicting fire-risk levels for the greater city of Montreal. Various datasets have been leveraged comprised of fire incidents, crimes, property assessments, demographics, weather conditions and districts all pertaining to the greater city of Montreal and its affiliated cities. Besides the weather and population datasets, Ville de Montréal publishes the datasets. The city of Montreal and its affiliated cities were split through square tesselation via a custom Python solution using geoPandas and Shapely. Each grid were rank by fire-risk for every month between 2015 and 2023. Decision tree models were used for data modelling where both random forest and XGBoost models performed well in accurately classifying high and low fire-risk areas, but XGBoost model was able to identify high-risk areas more efficiently.

Tools

A variety of tools and techniques were used in different steps of the project. Exploratory analysis was conducted using Python, Excel (pivot table), and Alteryx. Data cleaning, pre-processing, model development, and visualizations were primarily accomplished using Python (Jupyter notebook and libraries) and Tableau.

License

This codebase is a public domain, so feel free to use this repo for what you want.

About

A repository containning the assets for the data science cycle for predicting fire-risk levels for the greater city of Montreal. Six datasets were leveraged for this project including fire incidents, crimes, property assessments, demographics and weather conditions. Technologies used include Python libraries, Tableau and Excel.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published