基于LF算法改进的动态蚁群聚类算法

he ant colony algorithm which is based on the group behavior of the ant colony is one of the cluster analysis algorithms. Among these algorithms, LF algorithm is the standard ant colony algorithm based on the model that ant colony sort the young ants. This paper improves the LF algorithm and proposes IDLF algorithm. It mainly contains the following improvements of the IDLF algorithm: proposing three kinds of new moving rules to enhance the moving purpose of the ants; improving the termination condition of LF algorithm so that the algorithm can terminate automatically according to the actual clustering; adjusting the relevant parameters dynamically to enhance the self-adaptability of the algorithm; proposing the solution to deal with the situation of long-term loading or unloading actions of the ants. The results of the experiments demonstrate that the clustering efficiency and accuracy of IDLF algorithm are better than those of LF algorithm .