Computational Intelligence-Based Search of Entertaining Rules in the Space ofPredator/Prey Games

Computer games are becoming a primary source of entertainment due to modern day computers and high resolution graphics. Game developers are now getting interested to quantitatively measure the entertainment value of games. In addition to its contents, entertainment value also depends on the particular genre of the game. In this work we introduce a set of entertainment metrics for the predator/prey genre of games. Further, we employ the proposed metrics for automatic generation of entertaining games using an evolutionary algorithm. The evolutionary algorithm starts with an initial set of randomly generated games and is guided by the entertainment metrics towards more entertaining population of games. The results produced are counter-checked against the entertainment criteria of human by conducting a human user survey and a controller learning ability experiment. The proposed system serves as an expert system, based on computational intelligence techniques, for automatic generation of entertaining games.