How Context or Knowledge Can Benefit Healthcare Question Answering?

Healthcare question answering (HQA) is a challenging task as questions are generally non-factoid in nature. Traditional information retrieval techniques do not perform well on non-factoid questions. Recent neural question answering systems are reported to have performance gains over traditional methods. However, little attention has been given to HQA as datasets are generally too small to train a neural model from scratch. Recently, several systems have been proposed to learn context representations for HQA. Despite moderate progress, these systems have not been thoroughly compared with state-of-the-art neural models, and these neural models were tested only on self-created datasets. This makes it difficult for practitioners to decide which models should be used for various scenarios. To address the above challenges, we develop a new joint model to incorporate both context and knowledge embeddings into neural ranking architectures. First, we adapt context embedding pre-trained from large open-domain corpus to small healthcare datasets. Second, we learn knowledge embedding from knowledge graphs to provide external information for understanding non-factoid questions. To evaluate how context or knowledge embedding can benefit HQA, we adapt many state-of-the-art methods for general QA to HQA, by injecting the context or knowledge information only, or both of them. Extensive experiments are conducted to compare our approach with those adapted methods and current HQA systems. The results show that our approach achieves the state-of-the-art performance on both HealthQA and NFCorpus datasets.