A novel method for discovering frequent changing patterns from historical web access data

作者：Zhu; Zhiguo

来源：ICIC Express Letters, 2014, 8(9): 2443-2449.

• Dynamic features
• Frequent changing patterns
• Intelligent web
• Scalability and performance
• Scalable algorithms
• Synthetic datasets
• Web access pattern
• Web usage mining

Mining frequent Web access pattern is a very hot research issue in the field of Web Usage Mining. However, most of the existing techniques depend on the snapshot of Web access data and neglect their dynamic features. In this paper, we focus on mining the novel patterns by measuring the changing degree in the evolution of Web access data. Firstly, we employ the unordered tree structure to represent historical Web access data and then present an efficient and scalable algorithm to mine these frequent changing patterns. Finally, we conduct some experiments with a real dataset and four synthetic datasets. The results show that the proposed algorithm has better scalability and performance and the extracted novel knowledge can be helpful for many applications, such as intelligent Web advertisement and adaptive Website construction.

• 出版日期2014