Detecting Events in Networks

An event is something that happens or takes place at a certain moment of time. But then everything is happening and everything is changing, so every moment there is an event of a sort. The question is, how do we detect the events that have a certain relevance and what would those be. Alain Badiou, one of the most prominent philosophers, defines the event as something that re-casts ontology in a new light. That is, a real event that has the capacity to change things is the one that transforms the context. It’s not only unpredictable or rare, it is unimaginable using the field of probabilities we’re operating within.

To put this into a more practical perspective: what would be an event like this within a network? Clearly, a network itself is often a multiplicity of events recorded over a period of time: interactions between different agents, possibilities for encounters, records of movements in space. So an event among the events that transcends the possibilities inscribed into the already existing network would be something that creates a new topology that has not been available before.

In their recent research “A Radar for the Internet”, M. Latapy, C. Magnien and F. Ouédraogo propose a novel way for detecting events in the internet by identifying the ego-network of a node through tracetree algorithm. Going through several iterations over a period of time they gradually map the standard topology for a group of nodes. The way they propose to detect the outstanding events is by observing this resulting topology and waiting for any sort of deviations to occur. As soon as a cluster of interconnected nodes emerges, which has the linkages that do not follow the already existing observed paths, the event is detected.

Try InfraNodus Text Network Visualization Tool developed by Nodus Labs. You can use it to make sense of disjointed bits and pieces of information, get visual summaries for text documents, and generate insight for your research process: www.infranodus.com

This very interesting approach provides an elegant, scientifically sound explanation for Badiou’s point: an event in a network is something that creates a new topology, merges the nodes that have never been connected before, transcends the existing edges. This is a very powerful metaphor because it can be applied in anything: from language to social interactions. A social event, in this case, would be an event that creates new communities and builds bridges between groups and people who’s never met before. An event in thinking would be a concept that creates the new links across the concepts and domains of knowledge in novel ways.

On the internet people come and go, but we would like to stay in touch. If you like what you're reading, please, consider connecting to Nodus Labs on Facebook, Twitter and Patreon, so we can inform you about the latest updates and engage in a dialogue.