Integration of Multiple Fuzzy Matrix _SOFM of Web users access pattern mining

摘要

When users visit web, it has variety of evaluated factors, take these situations into account, structuring Multiple fuzzy matrix, using several associated SOFM neural network being connected to carry out the study on the every evaluated factors, more accurate and scientific in Web Mining user access patterns. Finally, using this algorithm in simulation experiment, to some extent, improved user access pattern mining WEB accuracy and operational efficiency; through the simulation results prove that the algorithm is a relatively good clustering algorithm.