Exceptional model mining

Leiden Repository

Exceptional model mining

Title: Exceptional model mining
Author: Duivesteijn, Wouter
Publisher: Leiden Institute of Advanced Computer Science (LIACS), Faculty of Science, Leiden University
Issue Date: 2013-09-17
Keywords: Exceptional model mining
Supervised local pattern mining
Regression
Bayesian networks
Subgroup discovery
Abstract: Finding subsets of a dataset that somehow deviate from the norm, i.e. where something interesting is going on, is a classical Data Mining task. In traditional local pattern mining methods, such deviations are measured in terms of a relatively high occurrence (frequent itemset mining), or an unusual distribution for one designated target attribute (subgroup discovery). These, however, do not encompass all forms of "interesting". To capture a more general notion of interestingness in subsets of a dataset, we develop Exceptional Model Mining (EMM). This is a supervised local pattern mining framework, where several target attributes are selected, and a model over these attributes is chosen to be the target concept. Then, subsets are sought on which this model is substantially different from the model on the whole dataset. For instance, we can find parts of the data where two target attributes have an unusual correlation, a classifier has a deviating predictive performance, or a Bayesian network fitted on several target attributes has an exceptional structure. We will discuss some real-world applications of EMM instances, including using the Bayesian network model to identify meteorological conditions under which food chains are displaced, and using a regression model to find the subset of households in the Chinese province of Hunan that do not follow the general economic law of demand.
Description: Promotor: J.N. Kok, Co-Promotor: A.J. Knobbe
With Summary in Dutch
Faculty: Faculteit der Wiskunde en Natuurwetenschappen
Citation: Duivesteijn, W., 2013, Doctoral Thesis, Leiden University
Sponsor: This research is supported by the Netherlands Organisation for Scientific Research (NWO) under project number 612.065.822 (Exceptional Model Mining).
Handle: http://hdl.handle.net/1887/21760
 

Files in this item

Description Size View
application/pdf Full Text 32.28Mb View/Open
text/html Links to published articles 26.74Kb View/Open
application/pdf Cover 307.4Kb View/Open
application/pdf Title Pages_Contents 226.9Kb View/Open
application/pdf Chapter 1 222.8Kb View/Open
application/pdf Chapter 2 267.2Kb View/Open
application/pdf Chapter 3 372.2Kb View/Open
application/pdf Chapter 4 331.1Kb View/Open
application/pdf Chapter 5 299.3Kb View/Open
application/pdf Chapter 6 2.392Mb View/Open
application/pdf Chapter 7 410.6Kb View/Open
application/pdf Chapter 8 342.0Kb View/Open
application/pdf Chapter 9 364.1Kb View/Open
application/pdf Chapter 10 205.2Kb View/Open
application/pdf References 215.0Kb View/Open
application/pdf Summary in Dutch 196.3Kb View/Open
application/pdf Summary in English 196.3Kb View/Open
application/pdf Acknowledgements_Curriculum Vitae 213.7Kb View/Open
application/pdf Propositions 164.5Kb View/Open

This item appears in the following Collection(s)