Web Relationship Mining Based on Extended Concept Lattice
基于扩展概念格的Web关系挖掘

The lack of an efficient Web organization and management mechanism has become the bottleneck of Web applications. To address the problem, this paper proposes an extended concept lattice model based on Web input and output messages for service relationship mining. Concept coverage functionality is created and introduced into normal concept lattice. Off-Line algorithms for Web relations including equivalence, replacement, and flow relations mining, as well as their on-line update algorithms, are given. Experiment on real Web registry center shows that the extended concept lattice model is a highly efficient organization mechanism for Web relationships. Real Web registry center automatically mines, is expected to provide intelligent support for service choices, optimizations and composition.