Multi-thresholds Clustering Objects in a Road Network

Threshold selection is an important topic and also a critical preprocessing step, which directly affects the accuracy of the clustering in a road network. This paper analyzes the necessity of multiple thresholds selection in a road network, extracts the similar nature of the objects, proposes firstly the scheme of multiple thresholds based on support vector regression (SVR) and improves on the existing algorithm. Performance analysis and experimental result show that the multiple thresholds scheme achieves high efficiency and accuracy for clustering objects based in a road network.