Abstract

Abstract—Network data are more difficult to analyze when they originate from different types of relationships. In online social networks like Facebook, for example, interactions related to friendship, kinship, business, interests, and other relationships may all be represented as catchall “friendships. ” Because several relations are mingled into one, the resulting networks exhibit relatively high and uniform density. As a consequence, the variation in positional differences and local cohesion may be too small for reliable analysis. We introduce a method to identify the essential relationships in networks representing social interactions. Our method is based on a novel concept of triadic cohesion that is motivated by Simmel’s concept of membership in social groups. We demonstrate that our Simmelian backbones are capable of extracting structure from Facebook interaction networks that makes them easy to visualize and analyze. Since all computations are local, the method can be restricted to partial networks such as ego networks, and scales to big data. I.