A fusion clustering algorithm and its application in route optimization

Data mining is a nontrivial process so that we can identify the effective, unknown, potentially useful and ultimately apprehensible pattern from databases. Clustering analysis is an important approach of data mining. This paper introduces a new concept of Dynamic Data Windows, and then puts forward a new fusion clustering algorithm with Dynamic Data Windows, the idea of k-means algorithm and density-based method. This new fusion clustering algorithm overcomes some disadvantages of traditional methods. Comparing with clustering based on density, integrated clustering analysis algorithm and clustering based on ANN, the new fusion clustering algorithm is more valuable in data mining. This new fusion clustering algorithm was used in Geographic Information System (GIS). Some analysis results show that the significant improvement to ship-routing design using the new fusion clustering algorithm with Dynamic Data Windows in database of GIS.