A Graph Enhanced BERT Model for Event Prediction
This work addresses event prediction problems for applications like narrative understanding, but it is incremental as it builds on existing graph-enhanced BERT methods.
The paper tackles the challenge of predicting subsequent events by addressing the sparsity of event graphs, which limits relational feature retrieval, and proposes a method to automatically build event graphs using a BERT model with a structured variable. Results show that this approach outperforms state-of-the-art baselines on script event prediction and story ending prediction tasks.
Predicting the subsequent event for an existing event context is an important but challenging task, as it requires understanding the underlying relationship between events. Previous methods propose to retrieve relational features from event graph to enhance the modeling of event correlation. However, the sparsity of event graph may restrict the acquisition of relevant graph information, and hence influence the model performance. To address this issue, we consider automatically building of event graph using a BERT model. To this end, we incorporate an additional structured variable into BERT to learn to predict the event connections in the training process. Hence, in the test process, the connection relationship for unseen events can be predicted by the structured variable. Results on two event prediction tasks: script event prediction and story ending prediction, show that our approach can outperform state-of-the-art baseline methods.