This repository contains the code and resources for the "Traffic Severity Index Prediction" Project, conducted during the Monsoon semester of 2023.
The goal of this project is to predict the severity of traffic based on various factors such as weather conditions, road conditions, time of day, and more. This is particularly relevant to urban planning and public safety. By accurately predicting traffic severity, we can optimize traffic flow, reduce congestion, and potentially prevent accidents. This could lead to more efficient transportation systems, improved road safety, and even reduced environmental impact due to less idle time on the roads. Furthermore, this information could be invaluable to emergency services, enabling quicker response times in critical situations. Ultimately, the success of this project could significantly enhance the quality of life for all road users. .
- Data Preprocessing: The project includes scripts for cleaning and preprocessing the accident data, handling missing values, outliers, and more.
- Feature Engineering: The scripts for feature engineering help in creating meaningful features that can improve the performance of the machine learning model.
- Model Training and Evaluation: The project uses various machine learning algorithms to train models on the preprocessed data. It also includes scripts for evaluating the performance of these models using appropriate metrics.
- Prediction: The final part of the project involves using the trained model to predict the severity index of traffic. Range is from 1 to 4 where 1 means least severe and 4 means most severe
To use this project, clone the repository and run the desired model.
This project uses a variety of machine learning algorithms for prediction. Below is a brief description of each:
-
Random Forest: An ensemble learning method that operates by constructing multiple decision trees at training time.
-
Decision Tree: A flowchart-like model, where each internal node denotes a test on an attribute, each branch represents the outcome of a test, and each leaf node holds a class label.
-
XGBoost: An optimized distributed gradient boosting library designed to be highly efficient, flexible, and portable.
-
Gradient Boosting Algorithm: A machine learning technique for regression and classification problems, which produces a prediction model in the form of an ensemble of weak prediction models.
-
AdaBoost (Adaptive Boosting): A boosting algorithm used as an ensemble method to improve the stability and accuracy of machine learning algorithms.
-
MLP (Multi-Layer Perceptron): A type of artificial neural network that consists of at least three layers of nodes and is used for classification and regression.
-
Mixed Naive Bayes: A variant of Naive Bayes that makes an assumption of independence among predictors.
-
SVM (Support Vector Machines): A type of supervised machine learning model used for classification and regression analysis.
-
Logistic Regression: A statistical model used to model a binary dependent variable.
Please note that this project is for educational purposes and should not be used as a substitute for professional advice or services related to road safety or accident prediction.