Nowadays, significant advancement has been done for vehicles and urban roads for the purpose of safety, but still many accidents of high severities happen in urban roads. Factors related to the human being have uncertainty and are complex in nature and it is complicated to separate every feature since these show irregularity. Driver’s age, gender, driving experience, and education may be selected as human factors. Determination of the significant contributing factors is the primary task in road accident analysis. In view of that, this paper deals with the model of Adaptive Neuro-Fuzzy Inference System (ANFIS) is applied in predicting severity of accidents applying in urban roads. K-means clustering was applied, and the mean of silhouette value was selected as an indicator to identify the number of subgroups of the numerical variable for ANFIS model. Classification accuracy, specificity sensitivity, and precision values were being calculated for evaluating performance of ANFIS method. Here, the best-fit model was also selected according to mean square errors (MSE)). The classification accuracy reached 76.92% which reflects the high fitness due to the good classification accuracy which concludes that the model is suitable for prediction of accident severity as well as other classification-related applications.
[1]
F Mannering,et al.
Analysis of injury severity and vehicle occupancy in truck- and non-truck-involved accidents.
,
1999,
Accident; analysis and prevention.
[2]
P. Rousseeuw.
Silhouettes: a graphical aid to the interpretation and validation of cluster analysis
,
1987
.
[3]
Lai Zheng,et al.
Approach to Predict Road Accident Frequencies: Application of Fuzzy Neural Network
,
2011
.
[4]
J.-S.R. Jang,et al.
Input selection for ANFIS learning
,
1996,
Proceedings of IEEE 5th International Fuzzy Systems.
[5]
Jyh-Shing Roger Jang,et al.
ANFIS: adaptive-network-based fuzzy inference system
,
1993,
IEEE Trans. Syst. Man Cybern..
[6]
Gadadhar Sahoo,et al.
Feature selection in accident data: an analysis of its application in classification algorithms
,
2016,
Int. J. Data Anal. Tech. Strateg..
[7]
Mojtaba Ziyadi,et al.
Prediction of accident severity using artificial neural networks
,
2011
.
[8]
M. Sugeno,et al.
Derivation of Fuzzy Control Rules from Human Operator's Control Actions
,
1983
.