When Genetic Programming is used to evolve decision trees for data classification, search spaces tend to become extremely large. We present several methods using techniques from the field of machine learning to refine and thereby reduce the search space sizes for decision tree evolvers. We will show that these refinement methods improve the classification performance of our algorithms.

We show how the understandability and speed of genetic programming classification algorithms can be improved, without affecting the classification accuracy. By analyzing the decision trees evolved we can remove the unessential parts, called introns, from the discovered decision trees. Since the resulting trees contain only useful information they are smaller and easier to understand. Moreover, by using these pruned decision trees in a fitness cache we can significantly reduce the number of unnecessary fitness calculations.

Given a database, structure mining algorithms search for substructures that satisfy constraints such as minimum frequency, minimum confidence, minimum interest and maximum frequency. Examples of substructures include graphs, trees and paths. For these substructures many mining algorithms have been proposed. In order to make graph mining more efficient, we investigate the use of the ``quickstart principle'', which is based on the fact that these classes of structures are contained in each other, thus allowing for the development of structure mining algorithms that split the search into steps of increasing complexity. We introduce the GrAph/Sequence/Tree extractiON (Gaston) algorithm that implements this idea by searching first for frequent paths, then frequent free trees and finally cyclic graphs. We investigate two alternatives for computing the frequency of structures and present experimental results to relate these alternatives.

Molecular fragment mining is a promising approach for discovering novel fragments for drugs. We in-vestigate a method for mining fragments which consists of three phases: first, a preprocessing phase for turning molec-ular databases into graph databases; second, the Gaston frequent graph mining phase for mining frequent paths, free trees and cyclic graphs; and third, a postprocessing phase in which redundant frequent fragments are removed. We will devote most of our attention to the frequent graph mining phase, as this phase is computationally the most demanding, but will also look at the other phases.

Given a database of graphs, structure mining algorithms search for all substructures that satisfy constraints such as minimum frequency, minimum confidence, minimum interest and maximum frequency. In order to make frequent subgraph mining more efficient, we propose to search with steps of increasing complexity. We present the GrAph/Sequence/Tree extractiON (Gaston) tool that implements this idea by searching first for frequent paths, then frequent free trees and finally cyclic graphs. We give results on large molecular databases.