KETEPATAN KLASIFIKASI TINGKAT KEPARAHAN KORBAN KECELAKAAN LALU LINTAS MENGGUNAKAN METODE REGRESI LOGISTIK ORDINAL DAN FUZZY K-NEAREST NEIGHBOR IN EVERY CLASS

Published: 22 Jul 2015.
Open Access

Citation Format:
Abstract

Traffic accident is an accidental event on the road involving vehicles with or without another road users which causes damage for the victims. Semarang has quite high number of traffic accidents, which in 2014 occured 801 cases of traffic accidents. Based on the government regulation number 43 of 1993 about highway infrastructure and traffic, the impact of traffic accidents can be classified based on victims conditions such as minor injuries, serious injuries, and died. In this research will discuss about the accuracy of severity traffic accidents victim classification in Semarang in 2014 using Ordinal Logistic Regression method and Fuzzy K-Nearest Neighbor in Every Class (FK-NNC). The result of Ordinal Logistic Regression method analysis produces the accuracy of classification value is 90,5405%, meanwhile Fuzzy K-Nearest Neighbor in Every Class method produces the accuracy of classification method is 89,19%.

 

Keywords:      Traffic accidents, Ordinal Logistic Regression, Fuzzy K-Nearest Neighbor in Every Class

Keywords: Traffic accidents; Ordinal Logistic Regression; Fuzzy K-Nearest Neighbor in Every Class

Article Metrics: