Professor Mannila has worked on different aspects of pattern discovery in data mining. He and co-authors were among the first to find efficient algorithms for association rules. He has long been working in the area of pattern discovery from temporal data: the work of his team on finding episodes from sequences has attracted wide attention (thousands of citations in both SCI and on Google Scholar). He has also studied in detail the problem of sequence segmentation.
Professor Mannila's research is characterized by the interplay of algorithmic considerations and practical applications, especially the use of data mining methods in other sciences. He has cooperated with researchers in medical genetics, paleoecology, paleontology, and linguistics, among others. Publication forums include a wide variety of journals, from PNAS and Paleoecology to IEEE TKDE and ACM TODS. Recently, his main interest has been in the development of randomization methods for assessing the significance of data mining results.
Heikki Mannila has held positions at the University of Helsinki and Helsinki University of Technology, and has been a senior researcher at Microsoft Research in Redmond, WA, as well as a research fellow at Nokia Research in Helsinki. He has just been appointed Vice President for Research and Education of Aalto University, a new university in Helsinki formed as the merger of Helsinki University of Technology, Helsinki School of Economics, and the University of Art and Design Helsinki.
2009 IEEE ICDM Nomination and Evaluation Committees