Computational Models for Story Plot Generation and Summarisation

Abstract

Story plot generation and summarisation are two open problems in natural language newlinegeneration and understanding. Initial approaches to story generation and summarisation relied newlineheavily on extensive knowledge engineering. To overcome this expensive dependency, many newlineresearchers have designed neural-based approaches. However, the current neural models have newlinetheir fair share of weaknesses. Their generated story lacks many characteristics like a central newlinetheme, purpose, etcetera. It would hardly engage any reader. Additionally, these models are very newlineopaque, hardly shed any light on the storytelling process, and are far from explainable AI. newlineThis thesis proposes a mathematical model of story plot generation and summarization. newlineWe hypothesise that a story is a series of states which requires certain conditions to be met newlinebefore the transition of the states. Therefore, we use Petri net as a causal backbone of the model. newlineWe populate its knowledge from open data sources like ConceptNet and use a transformer model newlinelike Comet to predict the knowledge for unknown cases. This way, we synergise the best of both newlineworlds, where we have transparency and control like conventional models, while we do not have newlineto engineer knowledge like the neural models manually. newlineIn addition to proposing the model, we formalise many aspects of storytelling, like newlineconsistency and coherence. We propose multiple approaches utilising the model and newlineformalisation and address many challenges in the literature, like avoiding repetitive stories and newlinehandling flat characters. newlineAt last, we evaluate the proposed approaches and present our findings.