09 avril 2020 – Nantes
Dynamic exploration of large graphs for community detection
The graph is a powerful data representation for business intelligence specialists and analysts, but visualizing a graph can be complex as the number of nodes and vertices increases. Community detection algorithms solve this problem by grouping nodes into communities in a hierarchical fashion. But the computation is computationally intensive and the limitations come when the graph is dynamic.
LumenAI has given life to a new algorithm able to refresh communities in real time in a graph evolving over time.
It has applications in social network analysis, intelligence services, fraud detection, etc.