Semi-Supervised Learning for Neural Keyphrase Generation
This addresses the data scarcity issue in keyphrase generation for resource-limited domains, though it is incremental as it builds on existing seq2seq models.
The paper tackles the problem of generating keyphrases for document summarization by proposing semi-supervised methods that use both labeled and unlabeled data, achieving performance improvements over a state-of-the-art model trained only on labeled data.
We study the problem of generating keyphrases that summarize the key points for a given document. While sequence-to-sequence (seq2seq) models have achieved remarkable performance on this task (Meng et al., 2017), model training often relies on large amounts of labeled data, which is only applicable to resource-rich domains. In this paper, we propose semi-supervised keyphrase generation methods by leveraging both labeled data and large-scale unlabeled samples for learning. Two strategies are proposed. First, unlabeled documents are first tagged with synthetic keyphrases obtained from unsupervised keyphrase extraction methods or a selflearning algorithm, and then combined with labeled samples for training. Furthermore, we investigate a multi-task learning framework to jointly learn to generate keyphrases as well as the titles of the articles. Experimental results show that our semi-supervised learning-based methods outperform a state-of-the-art model trained with labeled data only.