https://github.com/srosalino/determining_traffic_accident_severity_in_the_usa

Helping the authorities to better understand traffic problems and to establish public policies to minimize this issue, and for insurance companies to define their commercial policy
https://github.com/srosalino/determining_traffic_accident_severity_in_the_usa

data-cleaning-and-preprocessing data-engineering data-wrangling feature-engineering machine-learning

Last synced: 4 days ago
JSON representation

Helping the authorities to better understand traffic problems and to establish public policies to minimize this issue, and for insurance companies to define their commercial policy

• Host: GitHub
• URL: https://github.com/srosalino/determining_traffic_accident_severity_in_the_usa
• Owner: srosalino
• Created: 2023-09-15T15:43:21.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2024-07-12T12:03:39.000Z (4 months ago)
• Last Synced: 2024-07-12T13:51:17.013Z (4 months ago)
• Topics: data-cleaning-and-preprocessing, data-engineering, data-wrangling, feature-engineering, machine-learning
• Language: Jupyter Notebook
• Homepage:
• Size: 73.5 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# **Previous note**

The jupyter notebook file contains all detailed information and explanation regarding every operation mentioned throughout the report. The report (file: *group2_slides_presentation.pdf*), therefore, does not replace the notebook to fully understand the whole data analysis carried out.

## **Research Question**

“What are the key factors that determine a higher/lower traffic accident severity in the USA?” 

To answer this question, we will examine three sub questions concerning three major factors:

* To what extent do weather conditions impact the severity of traffic accidents?

* To what extent do time conditions impact the severity of traffic accidents?

* To what extent do infrastructure conditions impact the severity of traffic accidents?

Furthermore, we will study the accident by geographical distribution throughout the country by states and cities; by date (months and days of the week) where accidents are prevalent; by duration in hours; and by distance in traffic congestion.

This study will help the authorities to better understand traffic problems and to establish public policies to minimize this critical issue, and for example for insurance companies to define their commercial policy. If we can identify the patterns of how these serious accidents happen and the key factors, we might be 

able to implement well-informed actions and better allocate financial and human resources. 

## **Data Sources**

Moosavi, Sobhan. (2019). US Accidents(2016 - 2021), Version 6 (Dec 2021). Retrieved 12th of January 2023 from https://www.kaggle.com/datasets/sobhanmoosavi/usaccidents?datasetId=199387&sortBy=voteCount&search=severity.

Moosavi, Sobhan, Mohammad Hossein Samavatian, Srinivasan Parthasarathy, and Rajiv Ramnath. “A Countrywide Traffic Accident Dataset.”, 2019.

Moosavi, Sobhan, Mohammad Hossein Samavatian, Srinivasan Parthasarathy, Radu Teodorescu, and Rajiv Ramnath. "Accident Risk Prediction based on Heterogeneous Sparse Data: New Dataset and Insights." In proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic 

Information Systems, ACM, 2019. 

* File name: “US_Accidents_Dec21_updated.csv” 

* File format: CSV 

* Last access date: January 25, 2023 

* Website: https://www.kaggle.com/datasets/sobhanmoosavi/us-accidents

## **Full Presentation**

The full presentation, in video format, is present in the file '*group2_video_presentation.mp4*'.