Formal Concept Analysis for Qualitative Data Analysis over Triple Stores

Business Intelligence solutions provide different means like OLAP, data mining or case based reasoning to explore data. Standard BI means are usually based on mathematical statistics and provide a quantitative analysis of the data. In this paper, a qualitative approach based on a mathematical theory called ”Formal Concept Analysis” (FCA) is used instead. FCA allows clustering a given set of objects along attributes acting on the objects, hierarchically ordering those clusters, and finally visualizing the cluster hierarchy in so-called Hasse-diagrams. The approach in this paper is exemplified on a dataset of documents crawled from the SAP community network, which are persisted in a semantic triple store and evaluated with an existing FCA tool called ”ToscanaJ” which has been modified in order to retrieve its data from a triple store.