My main research interest is in Algorithmic Data Analysis. In
particular, I am interested in applying matrix and tensor factorizations over non-standard algebras – for example, Boolean or Tropical algebras – to data mining problems. Modelling data mining problems, such as subgraph discovery, as matrix factorization problems allows us to utilize existing work from these seemingly unrelated fields and gives novel insights when developing new methods.
My other main brach of research is in redescription mining. I am particularly interested on the applications of redescription mining to other fields of science, such as biology, material sciences, and political science. Increasing the applicability of redescription mining or matrix and tensor methods requires advances in interactive and visual data mining; my research on interaction and visualisation naturally connects to the above topics.
Pauli MiettinenGeneralized Matrix Factorizations as a
Unifying Framework for Pattern Set Mining: Complexity Beyond
Proc. 2015 European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD '15), LNCS vol. 9285,
[manuscript | pdf (Springer) | slides]
Dóra ErdősPauli MiettinenDiscovering Facts with Boolean Tensor Tucker Decomposition.
Proc. 2013 ACM International Conference on Infortmation and Knowledge Management (CIKM '13),
Ervina CerganiPauli MiettinenDiscovering Relations using Matrix Factorization Methods.
Proc. 2013 ACM International Conference on
Information and Knowledge Management (CIKM '13),
Jan RamonPauli MiettinenJilles VreekenDetecting Bicliques in GF[q].
Proc. 2013 European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD '13),
[pdf (Springer) | manuscript]
Pauli MiettinenFully Dynamic Quasi-Biclique Edge Covers
via Boolean Matrix Factorizations.
Proc. 1st ACM SIGMOD Workshop on
Dynamic Networks Management and Mining (DyNetMM '13),
Esther GalbrunPauli MiettinenA Case of Visual and Interactive Data Analysis: Geospatial Redescription Mining.
ECML PKDD '12 Workshop on Instant Interactive Data Mining (IID '12),
Esther GalbrunPauli MiettinenFrom Black and White to Full Colour:
Extending Redescription Mining Outside the Boolean World.Proc. 2011 SIAM International Conference
on Data Mining (SDM2011),546–557.
[journal version |
Arianna GalloPauli MiettinenHeikki MannilaFinding Subgroups having Several Descriptions:
Algorithms for Redescription Mining.Proc. SIAM International
Conference on Data Mining (SDM),334–345. [pdf (SIAM)]
Pauli MiettinenTaneli MielikäinenAristides GionisGautam DasHeikki MannilaThe Discrete Basis Problem.Knowledge discovery in databases: PKDD 2006
– 10th European conference on principles and practice of
knowledge discovery in databases, Berlin, Germany, September 2006,Lecture Notes in Artificial Intelligence,4213,
(PKDD Best Paper)
[journal version | manuscript |
Pauli MiettinenMatrix Decomposition Methods for Data
Mining: Computational Complexity and Algorithms.
Publications of Department of Computer Science, A-2009-4,
Department of Computer Science, University
(Ph.D. thesis, monograph).
Certificate of Recognition,
Dissertation Award, 2010.
Pauli MiettinenThe Discrete Basis Problem
Department of Computer Science, University of
Helsinki2006 (M.Sc. thesis).
Stefan NeumannRainer GemullaPauli MiettinenWhat You Will Gain By Rounding: Theory and Algorithms for Rounding Rank.
Saskia MetzlerStephan GünnemannPauli MiettinenHyperbolae Are No Hyperbole: Modelling Communities That Are Not Cliques.
[an error occurred while processing this directive]
Pauli MiettinenJilles VreekenMDL4BMF: Minimum Description Length for Boolean Matrix Factorization.
Research Report MPI-I-2012-5-001,
Max-Planck-Institut für Informatik
| source code]
Pauli MiettinenA review of Mathematical Tools for Data
Mining: Set Theory, Partial Orders, Combinatorics by Dan A.
Simovici and Chabane Djeraba.SIGACT News42(2),43–46. 10.1145/1998037.1998049