SI 301 Lecture Notes - Lecture 4: Homophily, Clustering Coefficient, Nodeb
Document Summary
Local clustering coefficient(special case: when node has no pairs of friends, you get 0/0. If a node has less than 2 edges, we define its lcc to be zero. Embeddedness of an edge a-b is the number of friends a and b have in common: what s the embeddedness of a-b, 2. A and b are both friends of e and c. People connected by edges of high embeddedness can more easily trust each other: why, misbehavior risks in network, would you rather be node a or node b in this network, node b: involved in edges with high embeddedness. High clustering coefficient: many strong ties and high trust among her friends, node a: involved in local bridges(edges with low embeddedness). Low clustering coefficient: many weak ties spanning different communities. Structure holes: gaps in a network that do not allow two sets of people to communicate with each o ther.