Normalized Hierarchical SVM
This work addresses hierarchical classification challenges in domains like text or image categorization, offering incremental improvements over existing structured SVM methods.
The paper tackled the problem of large-scale hierarchical classification by introducing normalized regularization and margin in structured SVMs, achieving state-of-the-art results where unnormalized methods previously failed to outperform flat models.
We present improved methods of using structured SVMs in a large-scale hierarchical classification problem, that is when labels are leaves, or sets of leaves, in a tree or a DAG. We examine the need to normalize both the regularization and the margin and show how doing so significantly improves performance, including allowing achieving state-of-the-art results where unnormalized structured SVMs do not perform better than flat models. We also describe a further extension of hierarchical SVMs that highlight the connection between hierarchical SVMs and matrix factorization models.