TaxoKnow: Taxonomy as Prior Knowledge in the Loss Function of Multi-class Classification
This work addresses the challenge of leveraging label hierarchies for better classification, particularly benefiting domains with structured taxonomies, but it is incremental as it builds on existing regularization methods.
The paper tackles the problem of improving multi-class classification by integrating hierarchical taxonomy as prior knowledge into the loss function, resulting in significant performance gains in semi-supervised and fully supervised settings on industrial and public datasets.
In this paper, we investigate the effectiveness of integrating a hierarchical taxonomy of labels as prior knowledge into the learning algorithm of a flat classifier. We introduce two methods to integrate the hierarchical taxonomy as an explicit regularizer into the loss function of learning algorithms. By reasoning on a hierarchical taxonomy, a neural network alleviates its output distributions over the classes, allowing conditioning on upper concepts for a minority class. We limit ourselves to the flat classification task and provide our experimental results on two industrial in-house datasets and two public benchmarks, RCV1 and Amazon product reviews. Our obtained results show the significant effect of a taxonomy in increasing the performance of a learner in semisupervised multi-class classification and the considerable results obtained in a fully supervised fashion.