Insert or Attach: Taxonomy Completion via Box Embedding
This work addresses taxonomy completion for knowledge organization, offering a novel method to improve accuracy in enriching taxonomies, though it is incremental in advancing embedding techniques.
The paper tackles the problem of taxonomy completion by proposing TaxBox, a framework that uses box embeddings to model asymmetric relations for inserting or attaching new concepts, resulting in average performance boosts of 6.7% in MRR, 34.9% in Hit@1, and 51.4% in Prec@1 over prior methods on real-world datasets.
Taxonomy completion, enriching existing taxonomies by inserting new concepts as parents or attaching them as children, has gained significant interest. Previous approaches embed concepts as vectors in Euclidean space, which makes it difficult to model asymmetric relations in taxonomy. In addition, they introduce pseudo-leaves to convert attachment cases into insertion cases, leading to an incorrect bias in network learning dominated by numerous pseudo-leaves. Addressing these, our framework, TaxBox, leverages box containment and center closeness to design two specialized geometric scorers within the box embedding space. These scorers are tailored for insertion and attachment operations and can effectively capture intrinsic relationships between concepts by optimizing on a granular box constraint loss. We employ a dynamic ranking loss mechanism to balance the scores from these scorers, allowing adaptive adjustments of insertion and attachment scores. Experiments on four real-world datasets show that TaxBox significantly outperforms previous methods, yielding substantial improvements over prior methods in real-world datasets, with average performance boosts of 6.7%, 34.9%, and 51.4% in MRR, Hit@1, and Prec@1, respectively.