Skip to content

Latest commit

 

History

History
145 lines (94 loc) · 5.65 KB

readme.md

File metadata and controls

145 lines (94 loc) · 5.65 KB

Investigating the Car Accidents In Chicago

Amazon EC2 Link - https://aritheanalyst.com/carcrash

Table Of Contents

Setup

  • Install the requirements and setup the development environment.

    pip3 install -r requirements.txt

  • Run the application.

    python3 app.py

  • Navigate to localhost:5000.

Purpose

The goal of this research is to analyze chicago car accident reports data in order to classify the primary cause of an accident and answer the following questions:

  • Q1 - What is the distribution of car accident causes?

  • Q2 - What regions do the most car accidents occur?

  • Q3 - What effect do external factors have on the amount of car crashes and car crashes with injuries?

    • The time of the days effect on car accidents.
    • The weather’s effect on car accidents.

Data Description

Data files used:

  • veh.csv
  • crashes.csv
  • ppl.csv
  • dx.csv

Question 1:

What is the distribution of car accident causes?

EDA

Distribution of car accidents's primary causes

The most deadly types of crashes leading in proportions are Turning at 19% and Angle at 13%. I’ll recommend focusing on these as they account for the most fatalities.

Conclusion

The most types of car accidents are Rear Ends accidents accounting for 30% of car crashes. Followed by Sideswipe Same Direction accidents accounting for 16% of car crashes.

Recommendation

The most deadly types of crashes leading in proportions are: Turning at 19% and Angle at 13%. I’ll recommend focusing on these by making better and seperate traffic signals for turning as they account for the most fatalities.

Question 2:

What regions do the most car accidents occur?

EDA

Distribution of car accidents with injuries by street

Conclusion

From the geo-map showing the proportion of injuries caused by car accidents on each street by their longitude and latitude positions. As we can see from the map, more accidents with injuries occur about south-east from downtown of chicago, more especially on the major streets.

Recommendation

I'll definitely recommend channeling more law enforcers and first aid responders towards the south central region of chicago in order to keep up with the high concentration of serious accidents that occur within these area.

Question 3:

What effect do external factors have on the amount of car crashes and car crashes with injuries?

EDA

Time & Weather's effect on amount of accidents that occur

Conclusion

From the histogram we can see most accidents occur in clear weather but more accidents with injuries occurs when there’s hail or when it snows. We can also see more accidents occur from 4pm to 8pm.

Recommendation

My recommendation based on this would be to employ more government trucks to salt the street when it snows to minimize the chances of having a fatal accident.

Model

Random Forest, X Boosting & LinearSVC classifiers where implimented after re-sampling with SMOTE since the dataset was heavily imbalanced and they all gave roughly the same results give or take 5%. So I opted to go with X Boosting classifier using PCA as its feature selection parameter. The features included where:

  • Driver’s Action
  • Driver’s Vision
  • Roadway Surface Condition
  • Device Condition
  • First Crash Type
  • Posted Speed Limit
  • Age
  • Physical Condition

Interpretion

From the models feature importance plot below, we can see the most influential features are:

  • Driver’s Action - Improper Backing.
  • First Crash Type - Animal.
  • Driver’s Action - Disregarded Control Devices.

The model gave a log loss of 12.5 which and accuracy of 64%. This means the amount the model penelizes for incorrect predictions 12.5 but it only predicted 64% of the primary causes of accidents accurately.

Its total recall is 64% which is the total amount of times the model classified the cause of an accident was a category correctly out of the total amount of times that category was indeed the cause.

Its total precision is 64% and this is the total amount of times the model classified the cause of an accident was a category correctly out of the total amount of the predictions made for that category.

The AUC score is 0.8 and this is the average probability that the model can classifiy each class correctly.

The main metrics we'll focus on for this would be recall and the AUC score as we won't want to incorrectly classify the cause of an accident and risk wrongfully blaming a driver for an accident when it might have been something else.

Future Work

  • Road Type Division: Segregate the different types of streets/roads to understand the unique properties of accidents that occurs in each

  • More Data: Gather more data like if a driver was on the phone, exceeded the posted speed limit or or has a good amount of driving experience .

  • Region Division: Deeper analysis on the primary causes of accidents in the North, South, East West and Central regions of the city.