How can data (and graph) mining techniques support research in information systems?

Summary form only given. Daily uses of information systems generate large volumes of digital traces: queries on search engines, messages sent on Twitter, purchases on the Internet, new contacts in online social networks... Users sometimes leave traces without even noticing! These digital traces represent an extremely valuable source of information, provided that actual knowledge is extracted from them. In particular, the design and operation of the underlying information systems could be optimized in many ways, e.g., through personalization based on inferred user profiles, context-aware service recommendation, efficient resource allocation, process model extraction, etc.In this keynote, we will give an overview of data (and graph) mining techniques that can (should!) be used to analyze digital traces generated by information systems. Data mining is widely used in many areas, such as biology, marketing, finance and security; we will study and illustrate its potential to support research in the Information System domain.