The 20 revised long papers and the 19 revised regular papers presented were carefully reviewed and selected from 80 submissions. The papers are organized in topical sections on pattern mining, classification, outlier detection, clustering, feature construction and generation, knowledge acquisition, discovery science in reality, machine learning algorithms, Web mining, applications of predictive methods, and interdisciplinary approaches.

This volume contains the papers presented at the 7th International Conference on Discovery Science (DS 2004) held at the University of Padova, Padova, Italy, during 2–5 October 2004.

The main objective of the discovery science (DS) conference series is to provide an open forum for intensive discussions and the exchange of new information among researchers working in the area of discovery science. It has become a good custom over the years that the DS conference is held in parallel with the International Conference on Algorithmic Learning Theory (ALT). This co-location has been valuable for the DS conference in order to enjoy synergy between conferences devoted to the same objective of computational discovery but from different aspects. Continuing the good tradition, DS 2004 was co-located with the 15th ALT conference (ALT 2004) and was followed by the 11th Symposium on String Processing and Information Retrieval (SPIRE 2004). The agglomeration of the three international conferences together with the satellite meetings was called Dialogues 2004, in which we enjoyed fruitful interaction among researchers and practitioners working in various fields of computational discovery. The proceedings of ALT 2004 and SPIRE 2004 were published as volume 3244 of the LNAI series and volume 3246 of the LNCS series, respectively.