Research On Ant Colony Clustering Algorithm Based On LF

Data mining is the technology which can dig hidden,potential,useful information and knowledge from massive data-As an active research topic in the field of data mining,clustering is an important means to analyze data and find useful information.Among the existing clustering algorithms,ant colony algorithm is the product of the application of swarm intelligence algorithm in clustering.It is based on two kinds of colony behavior:the accumulation of dead ants and the search for food sources.LF clustering algorithm is a behavior model generated from the accumulation of corpses.Ants will pile up dead ants.This paper is base on the traditional LF algorithm.it discussed the shortage of LFalgorithms firstly,then improved the shortcomings of the traditional LF algorithmsThe main work is as follows:1.Summarized The improvement and application of ant colony clustering algorithm in recent years.bricfly introducing two basic models of ant colony clustering algorithm,and compares the advantages and disadvantages of the two models,which lays a theoretical foundation for the improved algorithm.2.The traditional LF ant colony clustering algorithm has been studied,such as slow convergence rate,waste of resources caused by empty ant no-load and easy to get into local optimum.In order to improve the convergence rate,The principle of direct allocation is adopted in the initial stage of the algorithm,putting the ants on the data point at random and generating random global memory.The movement of loaded ants is guided by global memory during clustering,moving to the memory center which is using cosine similarity to determine the most similar.Global memory is updated after an iteration is complete.when the ant failed to pick up the data object,the principle of dissimilarity is adopted to move ants to the next data point In order to reduce the resource waste caused by the random movement of ants again.The results of experiments show that this algorithm improves the convergence speed greatly on the basis of ensuring the accuracy of the original algorithm3.In order to improve the LF algorithm with excessive parameter dependence and slow initial iteration speed,a new LF ant colony clustering algorithm with global memory of information entropy was proposed.The new algorithm introduces the concept of information entropy,which changes the rules of the original LF algorithm in which ants judge to pick up or drop data objects,and avoids the influence of random Numbers generated in the traditional rules on the efficiency and stability of the algorithm.Finally,different neighborhood radii are added in different periods of the algorithm operation.Our experiments show that the new algorithm can effectively reduce the parameter dependence and improve the stability of the algorithm.