Integrating Deep Event-Level and Script-Level Information for Script Event Prediction
This work addresses script event prediction for natural language processing, offering an incremental improvement by incorporating richer event semantics and multiple participant sequences.
The paper tackles script event prediction by integrating deep event-level and script-level information, proposing MCPredictor, a Transformer-based model that improves performance on the New York Times corpus.
Scripts are structured sequences of events together with the participants, which are extracted from the texts.Script event prediction aims to predict the subsequent event given the historical events in the script. Two kinds of information facilitate this task, namely, the event-level information and the script-level information. At the event level, existing studies view an event as a verb with its participants, while neglecting other useful properties, such as the state of the participants. At the script level, most existing studies only consider a single event sequence corresponding to one common protagonist. In this paper, we propose a Transformer-based model, called MCPredictor, which integrates deep event-level and script-level information for script event prediction. At the event level, MCPredictor utilizes the rich information in the text to obtain more comprehensive event semantic representations. At the script-level, it considers multiple event sequences corresponding to different participants of the subsequent event. The experimental results on the widely-used New York Times corpus demonstrate the effectiveness and superiority of the proposed model.