Event Representation Learning Enhanced with External Commonsense Knowledge
This work addresses a specific limitation in event representation learning for NLP researchers and practitioners, though it is incremental as it builds on prior methods by adding external knowledge.
The paper tackles the problem that event representations lack commonsense knowledge about participant intents and emotions, which limits their ability to distinguish subtly different events. By incorporating external commonsense knowledge about intent and sentiment, the model achieved 78% improvements on hard similarity tasks and better performance on script event prediction and stock market prediction.
Prior work has proposed effective methods to learn event representations that can capture syntactic and semantic information over text corpus, demonstrating their effectiveness for downstream tasks such as script event prediction. On the other hand, events extracted from raw texts lacks of commonsense knowledge, such as the intents and emotions of the event participants, which are useful for distinguishing event pairs when there are only subtle differences in their surface realizations. To address this issue, this paper proposes to leverage external commonsense knowledge about the intent and sentiment of the event. Experiments on three event-related tasks, i.e., event similarity, script event prediction and stock market prediction, show that our model obtains much better event embeddings for the tasks, achieving 78% improvements on hard similarity task, yielding more precise inferences on subsequent events under given contexts, and better accuracies in predicting the volatilities of the stock market.