A Review of Biomedical Event Trigger Word Detection

Trigger words refer to the characteristic words that can reveal the nature of the event, marking the occurrence of the event. Trigger word recognition is one of the sub-tasks of event extraction, which is an important part of information extraction. In recent years, with the continuous development of natural language processing in biomedical field, many scholars have developed various trigger word recognition methods. In this paper, we firstly introduce the background, concept, goal and significance of the trigger word detection, then classify the trigger word recognition methods, elaborate on advantages, disadvantages and applicable conditions of dictionary-based, rule-based, machine learning-based, deep learning based methods and some hybrid models. Finally, we analysis the problems and challenges of biomedical event trigger word detection, and look forward to the future research direction.