This project is developed to predict the severity of an accident given the current weather, road and visibility conditions in order to reduce the frequency of car accidents/collisions in a community. This model will be able to predict the severity of accidents/collisions, and alert the drivers to be more careful if the conditions are critical.
This project uses an updated dataset of collisions, provided by SPD and recorded by Traffic Records, to analyze how and what conditions would affect the severity of accidents. The dataset includes all types of collisions, display at the intersection or mid-block of a segment from 2004 to present, updated weekly.
The notebook illustrates the process of data prepration, analysis, machine learning, and evaluation.
The report summarizes the purpose, process, result and the conclusion of the project.
The presentation demonstrates and emphasizes the result, and conclusion of the project.
Thank you for reading!