Evolution of cooperativeness in a business game relying on acquaintance based trustworthiness assessment

Reputation systems have been popular in several online market places involving anonymous players as it can provide crucial information on the trustworthiness of an object or individual player to combat selfish and deceptive behaviors from peers. Individual feedbacks on the quality of past association are the fundamental building blocks of reputation systems. Careful consideration in aggregating feedbacks from different sources is in fact very important in computing a reliable value for trust worthiness to facilitate decision making in a social dilemma situation like that of online market places. In this paper we are considering a possible improvement to a reputation model like that of eBay, with our interest lying on investigating how the cooperativeness and population of cooperators would evolve if the weight of the feedback source was assigned on the basis of past association between players. We categorize the feedback source as direct source, gray source and opposition friendly source to define an aggregation method for trustworthiness assessment that considers applying a dynamically computed weight to each source of feedback. Our result shows that breaking feedback sources on the basis of acquaintance and assigning weight accordingly favors the evolution of cooperativeness in the player society as compared to models which do not classify the feedback sources. A genetic algorithm based spatial iterated prisoner's dilemma (SIPD) environment has been used to simulate the experiments.