蚁群算法在Web挖掘聚类系统的设计与实现教程方案教程方案.doc

基于蚁群算法的Web挖掘聚类系统的设计与实现 摘要 本文在研究现有Web挖掘中聚类技术的基础上，将一种改进的蚁群聚类算法应用于Web挖掘聚类系统中；并引入一种改进的蚁群算法应用于Web使用挖掘的用户事务聚类中。实验结果表明：与传统的聚类算法相比，基于蚂蚁的聚类算法在Web挖掘中具有一定的优势。 本文首先针对Web挖掘的过程进行概述，然后详细地分析了Web挖掘中聚类和分类技术存在的优缺点。深入讨论了几种改进的蚁群算法，在分析了现有算法应用于Web挖掘技术上的不足之后，提出一种改进后的聚类算法并应用于Web挖掘聚类系统中，重新定义构造蚂蚁的方式、相似度和行为，通过仿真实验来对比传统的算法，来证明改进的算法对于Web聚类挖掘的适用性。 最后，引入了一种改进的蚁群算法(Improved Ant Colony Algorithm, IACA),结合其聚类分析模型并对算法进行实现，将其应用到Web使用挖掘的聚类模型上。通过实验仿真，该聚类算法在聚类过程中，很好地解决了算法中停滞的问题，并且在全局优化方面表现出较好的性能。 关键词：Web挖掘；蚁群算法；Web内容挖掘；Web使用挖掘；聚类 ABSTRACT Based on the study of existing Web mining the clustering technology based on an improved ant clustering algorithm was applied to class Web mining clustering system; And introducing a kind of improved ant colony algorithm is applied to Web use mining user affairs in clustering. The experimental results show that: with the traditional clustering algorithm, based on the ant clustering algorithm in Web mining has certain advantage. Firstly Web mining process are summarized in this paper, and then analyzed in detail in the Web mining the clustering and classification technology and the advantages and disadvantages of the existing. Further discussed several improved ant colony algorithm, the analysis of the existing algorithm applied in Web mining technology after shortcomings, this paper puts forward a improved clustering algorithm and applied in Web mining cluster system, redefine structure ants way, the similarity and the behavior, through the simulation experiments to compared with the traditional algorithm, to prove that the improved algorithm for Web clustering mining applicability. Finally, we introduce a kind of Improved Ant Colony Algorithm (Improved Ant Colony Algorithm, IACA), combined with its clustering analysis model and the Algorithm realization, its application to Web use mining clustering model. Through the experimental simulation, this clustering algorithm in the clustering process, provides good solution algorithm stagnation problems, and in the aspects o