Evolutionary Learning Model of Social Networking Services with Diminishing Marginal Utility

We propose a model of a social networking service (SNS) with diminishing marginal utility in the framework of evolutionary computing and present our investigation on the effect of diminishing marginal utility on the dominant structure of strategies in all agents. SNSs such as Twitter and Facebook have been growing rapidly, but why they are prospering is unknown. SNSs have the characteristics of a public goods game because they are maintained by users posting many articles that incur some cost and because users can also be free riders, who just read articles. Thus, a number of studies aimed at understanding the conditions or mechanisms that keep social media thriving theoretically by introducing the meta-rewards game, which is a variation of a public goods game. The meta-rewards games assume constant marginal utility, meaning that the rewards by receiving comments increase linearly according to the number of comments, but describing the psychological rewards of humans is often inappropriate. In this paper, we present our modification of the model using the diminishing marginal utility and our comparison of the experimental results with those of the original meta-rewards game. We demonstrate that the structure of dominant strategies of all agents in our game is quite different from that in the original meta-rewards game and is more reasonable to explain the users' behavior in SNSs because their efforts in SNSs are limited even if they have many friends.