Identification of creativity in collaborative conversations based on the polyphonic model

Abstract

The paper presents a theoretical approach and a set of experiments that operationalize it for the identification of creative moments in conversations. State-of-the-art artificial intelligence technology is used for the operationalization: natural language processing, machine learning, and deep neural networks The approach is based on the polyphonic model introduced by Trausan-Matu, which starts from Mikhail Bakhtin's analogy of discourse building in texts with polyphonic music. The divergent and convergent steps of creativity are related to the inter-animation of voices through dissonances and consonances in polyphonic, contrapuntal music.