| ||||||||
| ||||||||
| ||||||||
|
Given a database of structures, an algorithm for mining structured data is an algorithm that searches for interesting information in these structures. Examples of structures are graphs, trees, molecules, XML documents and relational databases. At the moment this homepage concentrates on algorithms that search for descriptions of structures. Descriptions may be subgraphs, subtrees, `submolecules', etc., that satisfy constraints such as minimum frequency, minimum confidence, minimum interest, maximum frequency, etc. The target of descriptive mining algorithms is not to find a predictive model, but to find a good description of existing data in order to gain insight into that data. |