Towards a virtual peer that writes stories with children

• Main Authors: Chua, Hans Gabriel, Cu, Geraldine Elaine, Ibarrientos, Chester Paul, Paguiligan, Moira Denise, Ong, Ethel Chua Joy
• Format: text
• Published: Animo Repository 2017
• Subjects: Natural language generation (Computer science); Intelligent agents (Computer software); Computer Sciences
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/4293
• Institution: De La Salle University

Children encounter more difficulty in writing than in sharing their stories verbally. This difficulty may be due to the need for them to organize their thoughts and express them in the written language. A software agent can be designed to mimic the support given by a human teacher or peer, such as asking questions, giving directives, and recommending possible story text. We describe how these tasks can be achieved by our software agent, Alice, using a combination of computational text understanding and generation techniques. Children found Alice's prompts to be useful in situations when they were having a hard time writing their story. Those children who opted to include Alice's suggested story text did so because they found the peer to state certain facts that can add new details to their stories. © 2017 Asia-Pacific Society for Computers in Education. All rights reserved.