Designing Labeled Graph Classifiers by Exploiting the Rényi Entropy of the Dissimilarity Representation

Representing patterns by complex relational structures, such as labeled graphs, is becoming an increasingly common practice in the broad field of computational intelligence. Accordingly, a wide repertoire of pattern recognition tools, such as classifiers and knowledge discovery procedures, are nowadays available and tested for various labeled graph data types. However, the design of effective learning and mining procedures operating in the space of labeled graphs is still a challenging problem, especially from the computational complexity viewpoint. In this paper, we present a major improvement of a general-purpose graph classification system, which is conceived on an interplay among dissimilarity representation, clustering, information-theoretic techniques, and evolutionary optimization. The improvement focuses on a specific key subroutine of the system that performs the compression of the input data. We prove different theorems which are fundamental to the setting of such a compression operation. We demonstrate the effectiveness of the resulting classifier by benchmarking the developed variants on well-known datasets of labeled graphs, considering as distinct performance indicators the classification accuracy, the computing time, and the parsimony in terms of structural complexity of the synthesized classification model. Overall, the results show state-of-the-art standards in terms of test set accuracy, while achieving considerable reductions for what concerns the effective computing time and classification model complexity.