Welcome to our GitHub repository, where we are dedicated to enhancing road safety in the bustling metropolis of New York City. With a mission to safeguard pedestrians, cyclists, and motorists navigating the city's streets, our project presents a predictive model that evaluates the likelihood of road accidents.
Utilizing the comprehensive Motor Vehicles Collisions - Crashes dataset from NYC Open Data, we've crafted a model that sifts through a decade's worth of collision data. Our rigorous data cleaning process ensured the removal of duplicates, outliers, and inconsistencies, setting the stage for a robust analysis.
Through intricate data visualization techniques, including scatter plots, heat maps, and histograms, we've unearthed patterns and correlations that inform our predictive framework. This framework employs machine learning to anticipate collision probabilities, offering a valuable tool for city officials to allocate resources effectively and enhance public safety.
Our model's insights into accident likelihoods empower decision-makers to target high-risk areas, such as specific intersections, with appropriate safety measures. By predicting potential hotspots for accidents involving various road users, we contribute to creating a safer urban environment for all.
In summary, our project harnesses big data analytics to address the critical issue of road safety in NYC. By predicting collision likelihoods and identifying risk factors, we aim to support the city's efforts in reducing accidents and protecting its citizens.
Authors: Soham Faldu, Sheel Patel, and Ohm Patel | New York University
For inquiries or contributions, please contact us at: