How Studying Ant Behavior Can Make Social Networks Better

Illustration for article titled How Studying Ant Behavior Can Make Social Networks Better

People who go on social networking sites are often compared to ant colonies — but it may be truer than anyone realizes. A new study from the University of Madrid suggests that the behavior of real-life ants could inspire developers to improve the way these sites function. By constructing an algorithm modeled off the foraging behavior of ants, researchers are hoping to accelerate the way relationships are established in social networks — and these findings could have implications for the development of such things as GPS navigation and online gaming.


Social networking sites are becoming increasingly generalized on account of their explosive popularity. A consequent problem facing today's software developers is in locating the referential chain that leads from one person to another - or what the developers describe as the path from node to node. As these networks increase in size, so too does latency. And as any user of social networks can attest, delays are a complete turn-off.

This is where the ants come in.

Ants have evolved extremely sophisticated behaviors when it comes to looking for food. Their foraging techniques have been delicately refined through the processes of natural selection — and now, researchers want to tap into that biological insight in an effort to develop an algorithm that could vastly improve the efficiency of any kind of software that needs to make quick associations between related elements.

Illustration for article titled How Studying Ant Behavior Can Make Social Networks Better

Biologists know that ants utilize a kind of biological algorithm to perform a similar task when they're looking for food. Specifically, ants are capable of finding the path between the anthill and the source of food by secreting and following a chemical trail, called a pheromone, which is deposited on the ground. In the ant world, ants catch a sniff of other scented trails, allowing them to follow both the pheromone as well as the scent of food helping them to find the food more quickly.

By engaging in this form of biomimicry, the researchers have developed a similar algorithm, called SoSACO, that works by accelerating the search for routes between two nodes that belong to a graph that represents a social network. In other words, the software finds better and more relevant connections between elements. For ants, it's food; for social networking users, it's any kind of connection that's important to the experience, like shared interests between friends.

This research was led by Jessica Rivero in UC3M's Laboratorio de Bases de Datos Avanzadas (The Advanced Data Bases Laboratory - LABDA). The paper, titled "Using the ACO algorithm for path searches in social networks", appears in the journal Applied Intelligence.


The implications of Rivero's research extend beyond social networks. It is thought that SoSACO can be used to improve GPS navigation devices, delivery routes for freight trucks and couriers, and even online gaming.

Via. Top image via Shutterstock / megainarmy. Body image via UC3M.



From what I've seen, this type of effort leads to 'superhighway' creation, accelerated over-popularity of a specific type of resources, and any other number of issues that make a huge complex system appear overly simplistic and, to users, boring. The system learns to route around all the nooks and crannies where interesting things wait to be discovered and send people directly to things already deemed 'popular' by the masses, which usually leads to a short shelf-life and burnout. It also is very attractive to those that enjoy gaming a system.

It's very good at supporting zombie ad revenue, but deprives resources to things that originally made the system interesting.