A comprehensive discussion of the related work is given in our technical report [BN05b]. The two most related approaches are Bunch [MMCG99] and the work of Eick and Wills [EW93]. The novelty of CCVisu is twofold: First, it is based on the co-change graph, not on syntax-based models. (There are other approaches using change history, but not for clustering.) Second, it is based on an energy model that is designed for clustering layout.
© Dirk Beyer