A novel ant-based clustering algorithm with an artificial force field

作者：Zhang, Lei; Cao, Qixin

DOI：10.1109/ICNC.2010.5584546

• Ant-based clustering
• Ant-based clustering algorithm
• Artificial force fields
• Classic algorithm
• Clustering quality
• Computation efficiency
• Iterative cycles
• Principal Components

This paper presents a novel ant-based clustering algorithm. In the algorithm, the objects are first preprocessed by Principal Component Analysis(PCA), then their two principal components are retained and processed as the projecting coordinates. Moreover, different from conventional ant-based clustering algorithms in which the objects are picked up or dropped by virtual ants, our proposed algorithm looks each object as an ant. After the objects are projected to the plan, an artificial force field is created. The object (ant) is attracted by the similar and repelled by the dissimilar ones in its local surrounding. The object moves to a certain place at the work of these forces. The moving direction and moving range are determined by the composite of all forces. The clusters are created by this force influence after many iterative cycles. The paper gives the detailed process of the algorithm. The performance of the algorithm is compared with other classic algorithms on synthetic and real datasets. The results are very encouraging in terms of the computation efficiency and clustering quality.

• 出版日期2010
• 单位上海交通大学