Over the last decade, research in local pattern discovery has
developed rapidly, and a range of techniques is available for
producing extensive collections of patterns. Because of the exhaustive
nature of most techniques, the pattern collections provide a fairly
complete picture of the information content of the database. However,
in many cases this is where the process stops. The local patterns
represent fragmented knowledge, and often it is not clear how the
pieces of the puzzle can be combined into a global model, which is
often the expected result of a data mining process.

Consequently, the question of how to turn large collections of
patterns into global models has received considerable attention. In
data mining, the probably bestknown line of work that follows this
approach are various approaches for associative classification, such
as CBA and its successors. Similar problems also occur in
propositionalization approaches to relational learning, where the goal
is to discover propositional features that may be relevant for a
propositional global model, or in related areas such as
bioinformatics, text mining and computer vision, where the process of
discovering complex features may be viewed as a local pattern
discovery search.

In general, all these approaches follow three different phases:

1. Local Pattern Discovery: finding a collection of local patterns in
a given database that satisfy a set of inductive constraints, such as
frequency or correlation with the target variable.

2. Pattern Subset Selection: selecting from the set of discovered
patterns a small but informative subset of patterns that shows
little redundancy.

3. Global Modeling: turning the set of selected patterns into an
actionable global model by combining patterns effectively and
dealing with potential conflicts between the patterns found.

While each of these phases has been studied well in isolation, their
dependencies are not yet well understood. Research areas that are of
interest to this special issue include:

* Associative Classification
* Combination Strategies
* Compression-based Pattern Selection
* Constraint-based Pattern Set Mining
* Efficient Pattern Set Discovery
* Ensembles of Patterns
* Feature Construction and Selection
* Generalilty of Local Pattern Constraints
* Global Modeling with Patterns
* Iterative Local Pattern Discovery
* KDD Process-models for Building Global Models from Local Patterns
* Metrics for Pattern Set Selection
* Pattern Ordering
* Patterns and Information Theory
* Pattern Set Selection
* Pattern Teams
* Propagation of Global Modeling Constraints to Local Pattern
Discovery
* Propositionalisation
* Quality Measures for Pattern Sets
* Resolution of Conflicting Predictions
* Subgroup Discovery

The full text of the call for papers can be found at:
http://www.springer.com/cda/content/document/cda_downloaddocument/CFP_10618_20081111.pdf
You can also reach this page from the DMKD journal site of Springer
(http://www.springer.com/computer/database+management+%26+information+retrieval/journal/10618,
click on "Call for Papers - Special Issue on ...").

Please don't hesitate to contact us for any further questions.

Kind regards,

Arno Knobbe, Johannes Fuernkranz
Guest editors special issue