Skip to content

Used nearest neighbors classification to determine which parts of the United States would contain more severe accidents

Notifications You must be signed in to change notification settings

tonywu1999/US-Accidents

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 

Repository files navigation

US ACCIDENTS SEVERITY CLASSIFICATION

Co-Authors: Anthony Wu, Erlun Lian, Amy Liang, Caroline Hoerrner, Coco Cai

Run the code on team17.ipynb to see the results of our classification methods.

To see the visualization of our results, you can look at the image called "Map USA.png". 4 denotes the highest severity of accidents while 1 denotes the lowest severity of accidents. Regions are shaded from 1 through 4 based on their projected severity from our classifier. We calculated that our algorithm produced 82% accuracy in classifying severity of certain regions.

Data we used on US Accidents can be found on kaggle:

https://www.kaggle.com/sobhanmoosavi/us-accidents

About

Used nearest neighbors classification to determine which parts of the United States would contain more severe accidents

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published