Intelligent approaches to the analysis of large data sets continue to be of dramatic importance for many real world applications. Ranging from manufacturing, agriculture, finance and many other industrial and scientific areas, in the past few years increasingly large reservoirs of data of diverse type have been collected in the life sciences as well. In order to uncover the information hidden in these vast amounts of data, methods from different disciplines are required. Most prominently statistics and computer science but it has become increasingly clear that also detailed knowledge about the underlying domain is needed, especially in areas that attempt to analyse data from complex systems. The interaction between these disciplines with often very different vocabulary and the development of systems that interact with the user to find the desired answer(s) are still mostly open problems although in recent years encouraging progress has been made. To discuss this and similar issues, Xiaohui Liu (now with Brunel University, UK) established a series of symposia, the first being held in Baden-Baden, Germany in the summer of 1995, followed by London (1997), Amsterdam (1999), and Lisbon (2001). In 2003 the fifth symposium was held in Berlin, Germany. Over 180 papers were submitted, of which an international program committee helped to select 17 for oral and 38 for poster presentation. Afterwards the Organizing Committee of the conference selected 6 papers for this special issue. The authors were asked to revise and extend their original contributions and an additional round of reviews resulted in the six papers presented here.