The Research And Application Of The Distance-Based Ant Lion Optimizer

Swarm Intelligence optimization algorithm simulates the swarm intelligence behavior of gregarious displayed by mutual cooperation,which contains the interaction among the individuals,the individual and the environment,the individual and the group.As a simulation of biosocial,swarm intelligence optimization algorithm mainly uses local information to produce unpredictable group behavior,which is a kind of iterative optimization method,also a kind of evolutionary computation method.Ant Lion Optimizer is a swarm intelligence algorithm inspired by the hunting mechanism of antlions in nature.It is robust,easy to use and requires few control parameters.However,it cannot meet all of the needs as the study time of the algorithm is short and there are several drawbacks still need to be solved.So this paper is committed to its further research,trying to solve its existing problems and put forward improved algorithms.In addition,this paper explores its application in clustering.The main work and contribution are as follows:(1)A new Ant Lion Optimizer based on Distance,The Distance-Based Ant Lion Optimizer(DALO),is proposed.Considering the distance,when the ant walk randomly,DALO takes the distance factor and the fitness factor into account at the same time,and simultaneously update the ant colony position,randomize the ants's position.)By ensuring ants search around most antlions and enhancing the randomness,the algorithm can search as much space as possible,thus can get the global optimal value.Experiments show that The Distance Ant Lion Optimizer is more excellent in convergence,robustness and avoidance of local optimal performance than other Swarm Intelligence algorithms.(2)A new ALO clustering algorithm based on K-means is proposed,and ALO algorithm is used to determine the new center of the cluster,help K-means algorithm accelerates the clustering speed after determining each sample of the cluster.The sum of distance that the cluster center to all sample-positions is minimum when the cluster center is stable,and then clustering problem is transformed into the problem of searching k optimal cluster centers in the samples space.Experiments show that the ALO clustering algorithm based on K-means can get better clustering quality and fast convergence speed compared with other clustering methods.