Adding background knowledge to formal concept analysis via attribute dependency formulas

We present a way to add user's background knowledge to formal concept analysis. The type of background knowledge we deal with relates to relative importance of attributes in the input data. We introduce AD-formulas which represent this type of background knowledge. The background knowledge serves as a constraint. The main aim is to make extraction of clusters from the input data more focused by taking into account the background knowledge. Particularly, only clusters which are compatible with the background knowledge are extracted from data. As a result, the number of extracted clusters becomes smaller, leaving out non-interesting clusters. We present illustrative examples and results on entailment of background knowledge such as efficient testing of entailment and a complete systems of deduction rules. Categories and Subject Descriptors I.2.3 [Artificial Intelligence]: Knowledge Representation Formalisms and Methods--relation systems; H.2.8 [Database Management]: Database Applic...

• Type: Journal
• Year: 2008
• Where: SAC
• Authors: Radim Belohlávek, Vilém Vychodil
• Keywords: Applied Computing, Background Knowledge, Formal Concept Analysis, Input Data, SAC 2008