Emerging archetypes in massive artificial societies for literary purposes using genetic algorithms
This work addresses the challenge of automating fictional story creation for authors or AI systems, but it is incremental as it builds on existing genetic algorithm methods in a new domain.
The authors tackled the problem of generating coherent literary archetypes in simulated societies by using a genetic algorithm to determine the minimal number of character profiles needed for emergent scenes like 'natality control' and 'revenge', showing that parametrization is feasible and such archetypes can emerge.
The creation of fictional stories is a very complex task that usually implies a creative process where the author has to combine characters, conflicts and plots to create an engaging narrative. This work presents a simulated environment with hundreds of characters that allows the study of coherent and interesting literary archetypes (or behaviours), plots and sub-plots. We will use this environment to perform a study about the number of profiles (parameters that define the personality of a character) needed to create two emergent scenes of archetypes: "natality control" and "revenge". A Genetic Algorithm (GA) will be used to find the fittest number of profiles and parameter configuration that enables the existence of the desired archetypes (played by the characters without their explicit knowledge). The results show that parametrizing this complex system is possible and that these kind of archetypes can emerge in the given environment.