Title:
Grammar Correction for Event-to-Sentence

Thumbnail Image
Author(s)
Tien, Ethan
Authors
Advisor(s)
Riedl, Mark O.
Advisor(s)
Editor(s)
Associated Organization(s)
Organizational Unit
Organizational Unit
Supplementary to
Abstract
Automated story generation is the problem of generating words and sentences with the goal of telling a coherent story. To date, we have the Event-to-Event model that can generate more events that give us the building blocks for the sequence of events in a story. However, the difficulty lies in trying to translate these building blocks back into coherent sentences that tell a story. Although we have a baseline implementation of the Event-to-Sentence model, the results still end up being incoherent which leaves much room for improvement. We present a technique that takes advantage of a grammar correction model in order to fix the errors in the output and increase the comprehensibility.
Sponsor
Date Issued
2019-05
Extent
Resource Type
Text
Resource Subtype
Undergraduate Thesis
Rights Statement
Rights URI