Machine Learning - Toronto Fire Incidents
During my Data Science bootcamp, one of our projects leveraged Machine Learning to address a critical issue: detecting situations where fires pose a threat to people. Utilizing a dataset provided by Toronto Fire Services on Kaggle (https://www.kaggle.com/datasets/phyxle/toronto-fire-incidents), which included fire incident data from 2011 to 2016, we employed the Random Forest model to analyze and predict potential fire threats. Despite the vast amount of data available, we found it insufficient for making reliable predictions.
<br>Our analysis led to a significant conclusion and recommendation: the collection of data regarding existing fire suppression systems in buildings is crucial. Furthermore, we advocate for the promotion of having such systems, even if it's just a basic handheld extinguisher, as an essential safety measure. This project underscored the importance of comprehensive data in enhancing predictive models and the vital role of preventive measures in fire safety.