基于对称点距离的蚂蚁聚类算法

ANT CLUSTERING ALGORITHM BASED ON POINT SYMMETRY DISTANCE

针对待聚类的数据对象的对称性，提出了一种基于对称点距离的蚂蚁聚类算法．该算法不再采用Euclidean距离来计算类内对象的相似性，而是使用新的对称点距离来计算相似性．实验结果表明：与标准的蚂蚁聚类算法相比，该算法在处理带有对称性质的数据集时，可以更好的识别数据集的聚类数日和划分．

According to the symmetry of data sets, the ant clustering algorithm based on point symmetry distance was proposed. In this algorithm, assignment of points to different clusters is done based on point symmetry distance rather than the traditional Euclidean distance. Compared to standard ant clustering algorithm, the experimental results demonstrated it could defect the better number of clusters and the proper partitions from data sets when data sets possess the property of symmetry.