AI Stories: An Interactive Narrative System for Children
Project Overview
The document highlights the innovative use of generative AI in education, particularly through the development of AI Stories, an interactive narrative system designed for children. This system, which will be tested in pediatric wards over a three-year period, aims to merge education and play by enhancing language skills and creativity through dialogue. Utilizing computational creativity and natural language processing, AI Stories allows children to co-create narrative worlds, making the storytelling experience rich and engaging. By generating dialogue that resonates with children's interests and maintaining coherent storytelling, the system fosters language play and encourages imaginative expression. Overall, the initiative demonstrates the potential of generative AI to transform educational experiences by making learning interactive and enjoyable, thus supporting key developmental skills in young learners.
Key Applications
AI Stories - An interactive dialogue system that co-creates narratives with children.
Context: Applied in pediatric wards to support children's language development through storytelling.
Implementation: Developed using computational linguistics and natural language processing techniques, incorporating subsystems for dialogue generation.
Outcomes: Promotes language play, engages children in storytelling, enhances creativity and language skills.
Challenges: Ensuring coherent narratives, managing the vast possibilities of user input, and maintaining engagement over time.
Implementation Barriers
Technical barrier
The complexity of managing user input and generating coherent narratives from it.
Proposed Solutions: Utilizing subsystems to manage dialogue flow and employing reinforcement learning to optimize response selection.
Engagement barrier
Maintaining children's interest and ensuring the narrative remains engaging through humor and creativity.
Proposed Solutions: Incorporating humor and creativity into the dialogue generation to keep the interaction lively.
Project Team
Ben Burtenshaw
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Ben Burtenshaw
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai