Statistical foundations of data mining

From May 2014 this line of work is funded by an ERC Consolidator Grant called "Formalising Subjective Interestingness in Exploratory Data mining" (FORSIED).

From October 2009 until September 2013 this project was funded by the EPSRC (EP/G056447/1) and various other smaller grants.

My main goal is to develop solid statistical foundations for data mining (as different from machine learning). It is my belief that a critical challenge in doing this is tackling the notion of subjective interestingness of patterns. Much of our work is aimed at addressing this challenge either in general or in specific settings.

Publications: 
Title Authors Status Year
Interesting pattern mining in multi-relational data Eirini Spyropoulou, Tijl De Bie, Mario Boley journal paper 2014
Mining Approximate Multi-Relational Patterns Eirini Spyropoulou, Tijl De Bie conference paper 2014
Subjectively interesting alternative clusterings Kleanthis-Nikolaos Kontonasios, Tijl De Bie journal paper 2013
Maximum Entropy Models for Iteratively Identifying Subjectively Interesting Structure in Real-Valued Data Kleanthis-Nikolaos Kontonasios, Jilles Vreeken, Tijl De Bie conference paper 2013
A theoretical framework for exploratory data mining: recent insights and challenges ahead Tijl De Bie, Eirini Spyropoulou conference paper 2013
Subjective interestingness in exploratory data mining Tijl De Bie conference paper 2013
Mining Interesting Patterns in Multi-Relational Data with N-ary Relationships Eirini Spyropoulou, Tijl De Bie, Mario Boley conference paper 2013
Formalizing Complex Prior Information to Quantify Subjective Interestingness of Frequent pattern Sets Kleanthis-Nikolaos Kontonasios, Tijl De Bie conference paper 2012
Knowledge discovery interestingness measures based on unexpectedness Kleanthis-Nikolaos Kontonasios, Eirini Spyropoulou, Tijl De Bie journal paper 2012
Maximum Entropy Models and Subjective Interestingness: an Application to Tiles in Binary Databases Tijl De Bie journal paper 2011
An Information Theoretic Framework for Data Mining Tijl De Bie conference paper 2011
Subjectively Interesting Alternative Clusters Tijl De Bie conference paper 2011
Interesting Multi-Relational Patterns Eirini Spyropoulou, Tijl De Bie conference paper 2011
Maximum Entropy Modelling for Assessing Results on Real-Valued Data Akis Kontonasios, Jilles Vreeken, Tijl De Bie conference paper 2011
An information-theoretic approach to finding informative noisy tiles in binary databases Kleanthis-Nikolaos Kontonasios, Tijl De Bie conference paper 2010
A Framework for Mining Interesting Pattern Sets Tijl De Bie, Kleantis-Nikolaos Kontonasios, Eirini Spyropoulou conference paper 2010
ModuleDigger: an itemset mining framework for the detection of cis-regulatory modules Hong Sun, Tijl De Bie, Valerie Storms, Qiang Fu, Thomas Dhollander, Karen Lemmens, Mieke Verstuyf, Bart De Moor, Kathleen Marchal journal paper 2009
DISTILLER: a data integration framework to reveal condition dependency of complex regulons in Escherichia coli Karen Lemmens, Tijl De Bie, Thomas Dhollander, Sigrid De Keersmaecker, Inge Thijs, Geert Schoofs, Ami De Weerdt, Bart De Moor, Jozef Vanderleyden, Julio Collado-Vides, Kristof Engelen and Kathleen Marchal journal paper 2009
From Frequent Itemsets to Informative Patterns Arianna Gallo, Alessia Mammone, Tijl De Bie, Marco Turchi, Nello Cristianini technical report 2009
Explicit Probabilistic Models for Databases and Networks Tijl De Bie technical report 2009
Finding Interesting Itemsets using a Probabilistic Model for Binary Databases Tijl De Bie technical report 2009
The Condition-Dependent Transcriptional Network in Escherichia Coli Karen Lemmens, Tijl De Bie, Thomas Dhollander, Pieter Monsieurs, Bart De Moor, Julio Collado-Vides, Kristof Engelen, Kathleen Marchal journal paper 2009
MINI: Mining Informative Non-redundant Itemsets Arianna Gallo, Tijl De Bie, Nello Cristianini conference paper 2007
Discovering Transcriptional Modules from Motif, ChIP-chip and Microarray Data Tijl De Bie, Pieter Monsieurs, Kristof Engelen, Bart De Moor, Nello Cristianini, Kathleen Marchal conference paper 2005