The World Wide Web has used in daily routine lifestyle. It provides all information just like current news, product information, community blogs, and share market etc. Web usage mining is a challenging activity that aims to discover new, relevant and reliable information. In this paper, the authors propose an enhanced technique to the learning capabilities and reduce the computation intensity of a competitive learning multi-layered neural network using the three layers architecture. The proposed model uses three layer architecture with multilayer perceptron algorithm to discover and analyze useful knowledge from the available web log data.