A Pairwise Class Interaction Framework for Multilabel Classification

We present a general framework for multidimensional classification that captures the pairwise interactions between class variables. The pairwise class interactions are encoded using a collection of base classifiers (Phase 1), for which the class predictions are combined in a Markov random field that is subsequently used for multi-label inference (Phase 2); thus, the framework can be positioned between ensemble methods and label transformation-based approaches. Our proposal leads to a general framework supporting a wide range of base classifiers in the first phase as well as different inference methods in the second phase. We describe the basic framework and its main properties, including detailed experimental results based on a range of publicly available databases. By comparing the performance with other multilabel classifiers we see that the proposed classifier either outperforms or is competitive with the tested straw-men methods. We also analyse the scalability of our approach and discuss potential drawbacks and directions for future work.