Event Detection Using a Self-Constructed Dependency and Graph Convolution Network

The extant event detection models, which rely on dependency parsing, have exhibited commendable efficacy. However, for some long sentences with more words, the results of parsing are complex, because each word corresponds to a directed edge label. These edges do not all provide guidance model, and accuracy tools decreases increase in sentence length, resulting error propagation. To solve these problems, we developed an model that uses self-constructed graph convolution network. First, statistically analyzed ACE2005 corpus prune tree, combined named entity features generate undirected graph. Second, implemented enhanced network using multi-head attention mechanism understand representation nodes Finally, gating semantic structural information sentence, enabling us accomplish task. A series experiments conducted demonstrates proposed method enhances performance model.