Combining labeled and unlabeled data for biomédical event extraction

Abstract

In biomédical event extraction domain, there is a small amount of labeled data along with a large pool of unlabeled data. Many supervised learning algorithms for bio-event extraction have been affected by the data sparseness. In this paper, we present a new solution to perform biomédical event extraction from scientific documents, applying a semi-supervised approach to extract features from unlabeled data using labeled data features as a reference. This strategy is evaluated via experiments in which the data from the BioNLP2011 and PubMed are applied. To the best of our knowledge, it is the first time that the combination of labeled and unlabeled data are used for biomédical event extraction and our experimental results demonstrate the state-of-the-art performance in this task.