Accident Detection & Alert System Using Machine Learning

Globally, and especially in developing nations, the severity of road accidents is a major concern. The severity of a traffic collision may be lessened by understanding the major and supporting variables. As a result, a thorough investigation is needed to deal with this overwhelming problem. Using machine learning, this project explains how to analyse traffic incidents more thoroughly to gauge their severity. The severity of the accidents was influenced by various variables, including the weather, lighting, road surface, etc. In order to determine when and where accidents are most likely to happen, the random forest classification method and the logistic regression algorithm are both applied to a set of frequencies of highway locations accidents within a 24-hour period. The suggested model cautions users to drive cautiously in that area based on these predictions. 1.32 lakh individuals lost their lives in traffic accidents in 2020, the fewest in 11 years. The lowest number was 1.26 lakh in 2009. The number of road accidents decreased to 3.66 lakh last year, the smallest amount in the previous 20 years. The strong restrictions concealed in these widely used item sets usually expose the connections between the factors that influence accidents, which can be used to break them and reduce the frequency of accidents.