Research Reveals Varying Competition Levels in Network Growth, Challenging Traditional Models
New research proposes a statistical method to quantify the intensity of competition among nodes in growing networks, such as online news platforms and cinema attendance data. The study, which examines how new entries affect existing ones, found a spectrum of behaviors ranging from scenarios where new entrants have minimal impact to those where they significantly influence the system. Contrary to some traditional models like the Barabási-Albert model, which assume perfect competition for links, this research indicates that perfect competition is never observed in real-world systems. The study measures attention by tracking new interactions over consecutive intervals, using comments for news and ticket sales for movies. It introduces a 'slowing factor' to characterize the evolution of interactions, which typically decreases over time due to aging effects. This approach allows for a more nuanced understanding of how attention and activity are distributed within dynamic networks.
