Considering from the law of gravity, the influence between particles includes'distance and direction. After discussing the vector influence between data objects, it is applied in clustering algorithm. Vector influence function is presented from the scalar influence function and direction influence function. Two methods———similarity and sum are introduced to compute the direction influence. The algorithm deals with the core point by getting the projection of the points in its neighborhood to judge whether it is uniformity influence. Only uniformity influence points can be expanded to form clusters. The theoretical analysis and experimental results indicate that the algorithm can discover clusters with arbitrary shape and can effectively eliminate noise such as boundary sparse points. It solves the difficulties of clustering high dimensional spatial data such as the spatial distribution of the data,not obvious boundary between clusters, too many noise data points and the phenomenon that the distance between the nearest and farthest neighbors of a data point goes to zero etc. The algorithm improves the accuracy of clustering and offers better results of clnstering on various data sets. It executes effectively and efficiently. The algorithm is scalable and general. While transforming the semi-structure data into Euclid space,it will always appear boundary sparse objects,VICA can deal with the noise effectively. Therefore,the algorithm is proper with the high dimension data set,and also can be applied in the semi-structure data clustering.%从万有引力角度考虑,质点之间相互影响包括距离和方向两个方面.本文讨论数据之间的矢量感应,并将其应用于聚类算法ⅥCA.引入对象的标量感应函数和方向感应函数,提出矢量感应函数概念.并给出确定方向感应函数的两个方法:方向相似度法和累加法.将核心对象邻域中的对象投影,进行向量单位化,考察核心对象的邻域均匀感应程度,将与均匀感应核心对象均匀感应密度可达的对象聚成一个簇.理论分析和实验结果表明,算法可以处理任意形状的簇,有效地排除了稀疏感应对象这类噪声,并且可以解决高维数据聚类边界区分不明显、密度分布不均、类边界噪声对象多的问题,提高了聚类精度.由于感应函数是一个泛化定义,算法具有通用性和可扩展性.将半结构化数据变换到欧式空间时,容易出现边界稀疏对象,算法可以有效处理噪声.因此,算法适用于大规模的高维数据集合,也可用于半结构化数据聚类.
展开▼
