Multi-factor Influenced Integrated Driving Behavior Clustering Analysis Evaluation Model
Shaojie Liu, Ning Wang, Xiaoting Wang · 2019
For the discrete GPS driving behavior data, the traditional data clustering method cannot meet the rational clustering requirements. This paper analyzes and studies the commonly used clustering algorithms, as well as proposes an improved K-means++ algorithm for discrete data. In order to improve the accuracy of clustering, In order to improve the accuracy of clustering, the algorithm considers reducing the influence of non-principal components on data difference in the process of clustering difference metrics and gives the weighting process based on entropy weight method, so as to obtain a more reasonable clustering method. Finally, we implement the algorithm and apply it to the field of driving behavior analysis, and compare it with representative clustering algorithm. The experimental results show that the proposed algorithm has higher accuracy.