DBSCAN Algorithm Clustering for Massive AIS Data Based on the Hadoop Platform
Zhihua Chen, Jianming Guo, Qing Liu · 2017
With the rapid development of economy in central cities, Inland river traffic flow has grown rapidly, based on the vast amount of ship AIS information, there are a lot of inland river traffic characteristics. With the rapid growth of information and data, the size of AIS data has been as high as dozens of GB, and even can reach TB units, how to deal with such a huge amount of AIS data quickly is becoming more and more important. In this paper, aiming at the research of DBSCAN clustering of AIS data, and combining the Hadoop platform, a DBSACN clustering algorithm based on Hadoop platform for massive AIS data was put forward. the DBSCAN clustering algorithm is encapsulated under MapReduce parallel computing framework, through HDFS distributed storage and MapReduce distributed computing, which make full use of the advantages of Hadoop in dealing with big data and greatly improves the efficiency of the algorithm. The experimental results show that the proposed scheme can greatly improve the efficiency of DBSACN algorithm clustering processing AIS data and detect abnormal trajectory without reducing the quality of algorithm clustering.