Hierarchical Clustering Analysis Method Based on the Grid with Obstacle Space

The advantage of grid-based clustering method is its fast processing speed. The speed of clustering algorithm and the number of data objects is unrelated. To discover any size and shape of the clusterÿit is by the number of units on each dimension in the data space. In this method, the amount of data and computation time does not matter, calculations and data entry of the order does not matter, does not require the number of k-means algorithm to pre-specified cluster and so on. Clustering problem with obstacle constraints has very strong practical value in the spatial clustering analysis, and has become a research hotspot in recent years. Under the condition of existing obstacles constraints, the vast majority of the spatial clustering algorithm can’t effectively solve the problem of irregular obstructions. Thus it has a greater impact on the accuracy of the algorithm clustering results, and reduces the efficiency of the algorithm. To solve this problem, an obstacle constraint space grid-based hierarchical clustering algorithm, which is GSHCO algorithm, is proposed. The algorithm inherits the advantages of gridbased clustering algorithm, by defining the concept of barriers to grid to deal effectively with the obstacles of arbitrary shape, to achieve the purpose of found clusters of arbitrary shape; At the same time, the algorithm uses a hierarchical strategy which can effectively reduce the complexity of the algorithm with obstacles clustering and the algorithm is improved operating efficiency. The experimental results show that the GSHCO algorithm can deal with obstacles constrained clustering, and with higher performance and better clustering quality.