Towards the Creation of Learning Environments: Exploring the Potentialities of the Weighted Mutation Interactive Genetic Algorithm

摘要

The main research question that this paper deals with is the notion of control in algorithmic design and the possible ways through which a learning environment between the user and the computer can be established. In particular, emphasis is laid on Interactive Evolutionary processes, by presenting a Weighted Mutation Interactive Genetic Algorithm, followed by its thorough investigation and discussion. This computational structure provides the user with the capability of deeply integrating into a strictly defined evolutionary procedure, while guiding the morph genetic development with respect to personal and intuitive decisions. Reference is also made to Gordon Pask's Theory of Conversation, which appears to serve as a suitable conceptual tool to further explore the proposed algorithm's capacities. It is finally suggested that the presented process favors the creation of a symbiotic, circular relationship between the user and the computer, forming a scheme which can be potentially generalised and implemented in different, multiple ways.