Résumé
Formal Concept Analysis (FCA) provides an important approach in software reengineering for software understanding, design anomalies detection and correction. However, FCA-based approaches have two problems: (i) they produce lattices that must be interpreted by the user according to his/her understanding of the technique and different elements of the graph; and, (ii) the lattice can rapidly become so big that one is overwhelmed by the mass of information and possibilities. In this paper, we present a catalogue of important patterns in concept lattices, which can allow automating the task of lattice interpretation. The approach helps the reengineer to concentrate on the task of reengineering rather than understanding a complex lattice. We provide interpretation of these patterns in a generalized manner and illustrate them on various contexts constructed from program information of different open-source systems. We also present a tool that allows automated extraction of the patterns from concept lattices.