Multi-Task Learning of Keyphrase Boundary Classification
This work addresses a practical but underexplored task in natural language processing for scientific text analysis, with incremental improvements in performance.
The paper tackles the problem of keyphrase boundary classification in scientific articles, which is underexplored due to limited labelled data, by using multi-task learning with deep recurrent neural networks and achieves significantly better performance than previous state-of-the-art approaches, particularly for long keyphrases.
Keyphrase boundary classification (KBC) is the task of detecting keyphrases in scientific articles and labelling them with respect to predefined types. Although important in practice, this task is so far underexplored, partly due to the lack of labelled data. To overcome this, we explore several auxiliary tasks, including semantic super-sense tagging and identification of multi-word expressions, and cast the task as a multi-task learning problem with deep recurrent neural networks. Our multi-task models perform significantly better than previous state of the art approaches on two scientific KBC datasets, particularly for long keyphrases.