Applying Formal Concept Analysis on Intelligent Teaching Help System

The intelligent help system (IHS) is an important assistant platform, especially in online teaching system. At present most search methods of the help systems are mainly based on keyword matches in database query or hierarchical classifications. The outstanding problem is that users are required to have a certain ability of organizing Keywords, otherwise insufficient keywords would limit the query, resulting in low recall rate and excessive keywords introduce the problem of high false negative rate and return excessive information. Our approach is founded on the Galois lattices and on the automation of Formal Concept Analysis (FCA) theory. FCA provides a conceptual tool for the analysis of data (e.g. documentation) and for the formal representation of knowledge. We compared information retrieval based on concept lattices to database queries and to navigation in hierarchical classifications and concluded that the performance of lattice navigation is better than the use of hierarchical classification in given domain.