Prethodno priopćenje
https://doi.org/10.7307/ptt.v28i2.1643
A Multi-Classification Method of Improved SVM-based Information Fusion for Traffic Parameters Forecasting
Hongzhuan Zhao
; Key Laboratory of Cyber Physical Social Dependable Service Computation, Chongqing University; School of Automation of Chongqing University
Dihua Sun
; Key Laboratory of Cyber Physical Social Dependable Service Computation, Chongqing University; School of Automation of Chongqing University
Min Zhao
; Key Laboratory of Cyber Physical Social Dependable Service Computation; College of Computer of Chongqing University
Senlin Cheng
; Key Laboratory of Cyber Physical Social Dependable Service Computation, Chongqing University; School of Automation of Chongqing University
Sažetak
With the enrichment of perception method, modern transportation system has many physical objects whose states are influenced by many information factors so that it is a typical Cyber-Physical System (CPS). Thus, the traffic information is generally multi-sourced, heterogeneous and hierarchical. Existing research results show that the multi-sourced traffic information through accurately classifying in the process of information fusion can achieve better parameters forecasting performance. For solving the problem of traffic information accurately classification, via analyzing the characteristics of the multi-sourced traffic information and using redefined binary tree to overcome the shortcomings of the original SVM (Support Vector Machine) classification in information fusion, a multi-classification method using improved SVM in information fusion for traffic parameters forecasting is proposed. The experiment was conducted to examine the performance of the proposed scheme and the results reveal that the method can get more accurate and practical outcomes.
Ključne riječi
Cyber-physical system (CPS); Information fusion; Support vector machine (SVM); Multi-classification; Intelligent Transport System (ITS); Traffic parameters forecasting
Hrčak ID:
156724
URI
Datum izdavanja:
25.4.2016.
Posjeta: 1.065 *