A biomedical events extracted approach based on phrase structure tree

摘要

Biomedical events, known as fine-grained and complex relations, are very important to biomedical information extraction. Automatically extracting biomedical events from the literature has attracted significant interest in recent years. Biomedical Event Extraction (BEE) is a difficult and urgent problem in biomedical information extraction area due to the unstructured text and its lexical variations. This paper presents a BEE approach based on phrase structure tree consisting of text preprocessing, syntax parsing and event trigger recognition for the extracting the biomedical events. The experimental results show that our method performs well. Event extraction results are comparable to the state of the art system. We develop the biomedical event extraction system, which can be accessed by http://210.28.186.168:8080/bioEvent/.