The Effects of Message Sorting in the Diffusion of Information in Online Social Media

In this work, we propose an agent-based model to study the effects of message sorting on the diffusion of low- and high-quality information in online social networks. We investigate the case in which each piece of information has a numerical proxy representing its quality, and the higher the quality, the greater are the chances of being transmitted further in the network. The model allows us to study how sorting information in the agent's attention list according to their quality, node's influence and popularity affect the overall system's quality, diversity and discriminative power. We compare the three scenarios with a baseline model where the information is organized in a first-in first-out manner. Our results indicate that such an approach intensifies the exposure of high-quality information increasing the overall system's quality while preserving its diversity. However, it significantly decreases the system's discriminative power.