FORECASTING THE SEVERITY OF THE CONSEQUENCES OF ACCIDENTS WITH THE USE OF METHODS OF MACHINE LEARNING

This article discusses the possibility of developing a model for predicting traffic accidents on the basis of a crash database provided by the State Traffic Safety Inspectorate of Russia. An example of using the collected data to develop a model for predicting the severity of the consequences of an accident is given, and factors influencing this are analyzed.

1. Показатели состояния безопасности дорожного движения (Электронный ресурс - http://stat.gibdd.ru/).

2. Python Data Analysis Library (Элек-тронный ресурс – https://pandas.pydata.org/).

3. scikitlearn (Электронный ресурс – https://scikitlearn.org/stable/).

4. imbalanced-learn (Электронный ре-сурс – https://imbalanced-learn.readthedocs.io/en/stable/).

5. Haibo He, Edwardo A. Garcia “Learn-ing from Imbalanced Data”, IEEE Transac-tions on Knowledge and Data Engineering ( Volume: 21 , Issue: 9 , Sept. 2009 ) (Элек-тронный ресурс – https://ieeexplore.ieee.org/document/5128907/).

6. XGBoost documentation (Электрон-ный ресурс – https://xgboost.readthedocs.io/en/latest/).