Abstract

There has been a recent line of work automatically learning scripts from unstructured texts, by modeling narrative event chains. While the dominant approach group events using event pair relations, LSTMs have been used to encode full chains of narrative events. The latter has the advantage of learning long-range temporal orders, yet the former is more adaptive to partial orders. We propose a neural model that leverages the advantages of both methods, by using LSTM hidden states as features for event pair modelling. A dynamic memory network is utilized to automatically induce weights on existing events for inferring a subsequent event. Standard evaluation shows that our method significantly outperforms both methods above, giving the best results reported so far.

Highlights

  • Recurring sequences of events in prototypical scenarios, such as visiting a restaurant and driving to work, are a useful source of world knowledge

  • We propose a novel dynamic memory network model, which combines the advantages of both LSTM temporal order learning and traditional event pair coherence learning

  • Pairwise Mutual Information (PMI) is the co-occurrence based model of Chambers and Jurafsky (2008), who calculate event pair relations based on Pointwise Mutual Information (PMI), scoring each candidate event ec by the sum of PMI scores between the given events e0, e1, ..., en−1 and the candidate

Read more

Summary

Introduction

Recurring sequences of events in prototypical scenarios, such as visiting a restaurant and driving to work, are a useful source of world knowledge. Pichotta and Mooney (2016) experimented with LSTM for script learning, using an existing sequence of events to predict the probability of a event, which outperformed strong discrete baselines. LSTMs capture significantly more order information compared to the methods of Granroth-Wilding and Clark (2016), Rudinger et al (2015), and Modi (2016), which model the temporal order of only pairs of events. No direct comparisons have been reported between LSTM and various existing neural network methods that model event-pairs We make such comparisons using the same benchmark, finding that the method of Pichotta and Mooney (2016) does not necessarily outperform event-pair models, such as Granroth-Wilding and Clark (2016). Our code is released at https://github. com/wangzq870305/event_chain

Related Work
Problem Definition
Event Representation
Modeling Temporal Orders
Modeling Pairwise Event Relations
Training
Datasets
Hyper-parameters
Influence of Event Structure
Method
Influence of Network Configurations
Final Results
Conclusion
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.