A personalized webpage reconstructor based on concept lattice and association rules

Efficiency issue in automatic web-based information retrieval research becomes an important issue to users. Most knowledge and materials on the Internet is in either semi-structured or unstructured hypermedia form. When users found a webpage as searching result from the Internet for self-learning, sometimes they may not understand the meanings of specific part of the retrieved webpage easily. They need to spend a lot of time in finding more references manually to make they get clear idea of the original retrieved webpage told. If an agent can be developed to reconstruct the webpage that the users are browsing by inserting additional self-explainable documents' links at appropriate places in the original webpage, it will be perfect. This research uses formal concept analysis (FCA) and association rule methodology (ARM) to develop a Keyword Association Lattice (KAL). With the KAL, the webpages accessed by users can be analyzed and reconstructed automatically. A pedagogical software agent called K-Navi for users doing survey and self-learning on the Internet is developed.