Protein-Protein Interaction Extraction from Biomedical Literatures Based on a Combined Kernel

摘要

Automatically extracting protein-protein interaction (PPI) from biological literature is an important and challenging task in natural language processing (NLP). In this paper, we use an ensemble kernel to extract the PPI information. This ensemble kernel is composed with feature-based kernel and structure-based kernel using the parse tree of a sentence containing two protein names. Experiments conducted on the IEPA corpus show that this ensemble kernel is efficient at extracting protein-protein interaction information. The recall, precision and f-score on the IEPA corpus are 73.03%, 82.09% and 77.28% respectively, which outperform most of the state-of-the-art systems.