Mining Complex Data : ECML/PKDD 2007 Third International Workshop

This volume contains 20 papers selected for presentation at the Third Inter- tional Workshopon Mining Complex Data-MCD2007-held in Warsaw, Poland, September 17-21, 2007. MCD is a workshop series that started in conjunction with the 5th IEEE International Conference on Data Mining (ICDM) in Ho- ton, Texas, November27-30,2005.ThesecondMCDworkshopwasheldagainin conjunction with the ICDM Conference in Hong Kong, December 18-22, 2006. Data mining and knowledge discovery, as stated in their early de?nition, can today be considered as stable ?elds with numerous e?cient methods and studies that have been proposed to extract knowledge from data. Nevertheless, the famous golden nugget is still challenging. Actually, the context evolved since the ?rst de?nition of the KDD process, and knowledge now has to be extracted from data becoming more and more complex. In the frameworkof data mining, many softwaresolutions were developedfor theextractionofknowledgefromtabulardata(whicharetypicallyobtainedfrom relationaldatabases).Methodologicalextensionswereproposedtodealwithdata initiallyobtainedfromothersources, e.g., inthecontextofnaturallanguage(text mining) and image (image mining). KDD has thus evolved following a unimodal scheme instantiated according to the type of the underlying data (tabular data, text, images, etc.), which, at the end, always leads to working on the classical double entry tabular format. However, in a large number of application domains, this unimodal approach appears to be too restrictive. Consider for instance a corpus of medical ?les.