Paper Key : IRJ************854
Author: Ranjeet Yadav,Dakshita Joshi
Date Published: 02 Jan 2025
Abstract
The auto industry has made efforts to create safer vehicles, but traffic accidents are inevitable, as this review paper explains. If we develop accurate prediction models that are able to automatically classify the type of injury and severity of various traffic accidents, we might be able to identify patterns in dangerous crashes. These social and street mishap examples can be valuable to foster traffic security control approaches. We believe that measures should be based on scientific and objective surveys of the causes of accidents and the severity of injuries in order to achieve the greatest possible accident reduction effects with limited budgetary resources. The results of four machine learning paradigms used to model the severity of injuries sustained in traffic accidents are summarized in this paper. We considered brain networks prepared utilizing cross breed learning draws near, support vector machines, choice trees and a simultaneous half and half model including choice trees and brain organizations. The results of the experiment show that the hybrid decision tree-neural network approach performed better than the other individual machine learning paradigms.
DOI Requested