My research interests lie in the realms of Database Management Systems and Machine Learning. Currently, along with my advisor Prof. Minos Garofalakis, Prof. Joe Hellerstein, Prof. Michael Franklin and Daisy Wang, I work in the HeisenData project, a joint effort of UCB and Intel Research Berkeley, which aims towards the development of a Probabilistic Database System that treats the elements of uncertainty and probabilistic inference as first-class citizens of the system. The goals of the project are:

• to support a variety of statistical models that capture correlations among data items,

• to expose statistical modeling structures and inference algorithms to key DBMS components (e.g., query engine, query optimizer),

• to support a uniform, declarative means for higher-level applications to store, query, and learn from such probabilistic data.

Wang, D.Z., Michelakis, E., Garofalakis, M., and Hellerstein, J.M.: BayesStore: Managing Large, Uncertain Data Repositories with Probabilistic Graphical Models. Proceedings of VLDB 2008, 34th International Conference on Very Large Data Bases, Auckland, New Zealand, 24-30 August 2008.

Michelakis, E., Wang, D.Z., Garofalakis, M., and Hellerstein, J.M.: Granularity Conscious Modeling for Probabilistic Databases. Proceedings of the 1st Workshop on Data Mining of Uncertain Data, ICDM 2007, Seventh IEEE International Conference on Data Mining, Omaha, NE, USA, 28-31 October 2007.[ppt]

Garofalakis, M., Brown, K.P., Franklin, M.J., Hellerstein, J.M., Wang, D.Z., Michelakis, E., Tancau, L., Wu, E., Jeffery, S.R., and Aipperspach, R.: Probabilistic Data Management for Pervasive Computing: The Data Furnace Project. In IEEE Data Engineering Bulletin, Vol. 29, No. 1, March 2006 (Special Issue on Probabilistic Data Management), pp. 57-63.

Michelakis, E., Androutsopoulos, I., Paliouras, G., Sakkis, G., and Stamatopoulos, P., 2004: Filtron: A Learning-Based Anti-Spam Filter. In Proceedings of the 1st Conference on Email and Anti-Spam (CEAS 2004), Mountain View, CA, USA, 30-31 July 2004.

Androutsopoulos, I., Paliouras, G. and Michelakis, E., 2004: Learning to Filter Unsolicited Commercial E-Mail. N.C.S.R. "Demokritos", Technical Report, No. 2004/2, March 2004.

Michelakis, E., 2005: Active Classifier Learning from Positive and Unlabeled Examples. In Selected B.Sc. / M.Sc. Theses Book 2004-2005, Dept. of Informatics and Telecommunications, University of Athens (Master's Thesis, in Greek).

Michelakis, E., 2004: Automatic Filtering of Unsolicited Commercial E-Mail with Machine Learning Algorithms. In Selected B.Sc. / M.Sc. Theses Book 2003-2004, Dept. of Informatics and Telecommunications, University of Athens (B.Sc. Thesis, in Greek).