Based on the Improved K - means Algorithm of Tianjin Port Traffic Flow Characteristic Analysis

摘要

Mastering the characteristics of port ships traffic flow is the premise to plan, operate, and control waterway and navigation reasonably. Therefore, analysis of traffic flow characteristic of port ships is necessary and extremely important. Because Vessel traffic flow data acquisition methods are diverse, they contain flawed and incorrect data. To solve the problem, we made use of the improved K-means algorithm to prove the raw data of sequence samples. We applied data mining clustering analysis method to the Tianjin port ships traffic flow data characteristic for analysis. The results show that some valuable information is obtained by the proposed method which can also provide decision support for maritime safety administration.