accident-prediction

This is the implementation code for the paper, "A Dynamic Spatial-temporal Attention Network for Early Anticipation of Traffic Accidents"

Multi-dimensional Analytics Project on Road Accidents of India.

An analysis of traffic accident data for the UK in 2014, using data from the UK Data Service. (Sourced from Kaggle with original data coming from UK Data Service. See wiki for complete citations.)

Automatic detection and alert of vehicle accidents

ML web app for predicting motor vehicle accident severity using Karnataka State Police FIR data. Built for KSP Datathon 2024.

A Machine Learning approach to estimate the likelihood of road accidents based on road, weather, and environmental conditions

End-to-end Machine Learning pipeline for Road Accident Severity Prediction using US Accidents dataset, including EDA, preprocessing, model comparison, and optimization.

A hybrid AI model for predicting failures in water distribution systems using Adaptive Neuro-Fuzzy Inference System (ANFIS). The model integrates Genetic Algorithms (GA) and Ant Colony Optimization (ACO) to improve the accuracy of accident prediction.

An AI-Based Web Platform for Road Accident Risk Prediction Using Public Data

Smart road safety and accident risk prediction system using Next.js that analyzes traffic, weather, and historical data to visualize high-risk zones and suggest safer routes.

A machine learning pipeline to predict traffic accident severity. Uses SMOTE and Class Weights to tackle severe dataset imbalance with XGBoost and Neural Networks.

A machine learning pipeline to predict traffic accident severity. Uses SMOTE and Class Weights to tackle severe dataset imbalance with XGBoost and Neural Networks.

VisionZero is a machine learning-based accident severity prediction system that analyzes road accident data to identify risk factors and predict accident severity, supporting safer roads through data-driven insights

An end-to-end accident severity prediction project that predicts the seriousness of injury based on various features, using a multiclass classification model.

Chicagoland smart route planner: compare 8 routes by time, tolls, calm, and ML accident-risk (weather + crash trends). Streamlit, FastAPI PWA, and Expo.

Machine Learning-based system that predicts optimal ambulance deployment locations using accident data analysis and clustering techniques to reduce emergency response time and improve emergency healthcare services.

Source Code for Random Forest Accident Severity Prediction Modelling

SAFE: Spatial-temporal Adaptation in Federated Environments for CAV Accident Anticipation under Heterogeneity

Модель прогнозирования риска ДТП для каршеринга с F1 0.674 на основе нейронной сети. Использованы Python, PyTorch, CatBoost, PostgreSQL, Streamlit.

Vantage, AI-powered road safety intelligence platform for accident severity prediction, hotspot detection, road-risk analysis, and evidence-grounded infrastructure insights.