Enhancing event-based semantics in the ontology of picture books 2

• Main Author: Ang, Karen S.
• Format: text
• Language: English
• Published: Animo Repository 2012
• Subjects: Storytelling; Semantic computing; Natural language generation (Computer science)
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/4305
• Institution: De La Salle University

Automatic story generation systems must consider both consistency and coherency issues in the stories that they produce. However, this was not achieved in Picture Books 2 because its planning engine utilized a simple reasoning engine that retrieves the causal chain of events between the initial state and the target goal of the story from the manually-built storytelling ontology. In this research, we utilized different agent models - character agent, plot agent, and world agent - to enhance the planning process of Picture Books 2. Along with the agent models, the ontology structure utilized as the source of storytelling knowledge was also enhanced to support the tasks of the agents. Moreover, the system needed to track down the state of the world in each occurring event. Thus, a representation for the systems world states was developed. Results showed that the interaction among the agents enabled the story planner to perform better than the earlier version of Picture Books 2 in generating logical stories.