This paper presents a system based on detected information in social networks to support experiments in primate behavior psychology. The study aims to the learning processes through observation (e.g., usage of tools for breaking encapsulated fruits) and experiments that may emerge in an autonomous agents group (represented by capuchin monkeys). The experiment focuses on simulation data analysis for the prediction of social behavior associated with some cognitive skills. The dissemination of information occurs through the group social structure, which is represented by a graph object that has been built based on theorems and algorithms from the graph theory. The objective is to suggest items that may be of interest to the user. The finding of these contents uses techniques that analyze and combine data to reach the best possible degree of recommendation. These benefits coupled with the large volume of data focused on the right audience can boost business' profits or surveys. These benefits are the drivers for the high demand of such systems in different domains like films (e.g., NetFlix), social networks (e.g., Facebook), sales (e.g., IBM Watson), and others.