Leveraging External Knowledge for Out-Of-Vocabulary Entity Labeling
This addresses the challenge of handling unseen slots in real-world dialogue systems, which is an incremental improvement for the domain of natural language processing.
The paper tackles the problem of labeling out-of-vocabulary entities in multi-domain dialogue state tracking by introducing a neural network that leverages external knowledge bases to classify unseen slot keys and values, resulting in relative increases of 57.7% in F1 score and 82.7% in accuracy compared to previous methods.
Dealing with previously unseen slots is a challenging problem in a real-world multi-domain dialogue state tracking task. Other approaches rely on predefined mappings to generate candidate slot keys, as well as their associated values. This, however, may fail when the key, the value, or both, are not seen during training. To address this problem we introduce a neural network that leverages external knowledge bases (KBs) to better classify out-of-vocabulary slot keys and values. This network projects the slot into an attribute space derived from the KB, and, by leveraging similarities in this space, we propose candidate slot keys and values to the dialogue state tracker. We provide extensive experiments that demonstrate that our stratagem can improve upon a previous approach, which relies on predefined candidate mappings. In particular, we evaluate this approach by training a state-of-the-art model with candidates generated from our network, and obtained relative increases of 57.7% and 82.7% in F1 score and accuracy, respectively, for the aforementioned model, when compared to the current candidate generation strategy.