Nowadays, road safety has become an important issue in the urban areas due to the high vehicle density. Road safety can be improved by reducing the accidents. Road accident causes traffic hindrance which has become intolerable especially in big-cities. Therefore, analyzing the road accidents accurately can help to solve the problem of traffic crashes. In our project, we propose a hybrid model that combines both K-Nearest Neighbor and Support Vector Machines algorithm for road accident analysis and prediction of accident type, which is based on the hierarchical-learning approach. The accident types are classified as crash, drunk & drive, fire and skid. Our proposed model uses the combination of both KNN and SVM algorithms with the historical datasets collected from UCI Repository. This analyzed data will be more useful to suggest better safety measures to avoid traffic crashes. We experimentally analyze the performance of both KNN and SVM algorithms using R programming with large accident datasets. Results show that our hybrid model enhances the accuracy of road accident analysis.
Using hybrid Data Mining algorithm for Analysing road accidents Data Set
01.02.2019
218883 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Study on Road Accidents Using Data Mining Technology
IEEE | 2018
|Analysis of Road Accidents Using Data Mining Paradigm
Springer Verlag | 2021
|Analysing accidents by computer
Online Contents | 1995
Data mining on road safety: factor assessment on vehicle accidents using classification models
Taylor & Francis Verlag | 2016
|