Abstract | ||
---|---|---|
This paper addresses missing edges and vertices in a network. We discuss interchangeability and duality between vertices and edges in a graph. We use covariate information associated with vertices to estimate the probability of missing edges; likewise, we use covariate information associated with edges to estimate the probability of missing vertices. In order to predict missing vertices, we apply the line graph transformation, which converts edges to vertices and vertices to edges. The probability of an edge is obtained by taking the inner product of the vectors of covariates. Moreover, we have extended the methodology of predicting two edges (dyadic ties) to predict edges in a triad. The method is based on geometry and fuzzy logic. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/WI-IAT.2010.317 | WI-IAT), 2010 IEEE/WIC/ACM International Conference |
Keywords | Field | DocType |
missing edge,inner product,predicting edges,dyadic tie,missing vertex,line graph transformation,covariate information,duality,duality mathematics,geometry,computational modeling,graph theory,fuzzy logic,markov processes,social networks,line graph,probability,spreading activation,estimation theory,network,data models | Graph center,Data mining,Path (graph theory),Computer science,Mixed graph,Artificial intelligence,Multiple edges,Path graph,Combinatorics,Distance,Cycle graph,Machine learning,Topological graph | Conference |
Volume | ISBN | Citations |
3 | 978-0-7695-4191-4 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Walid K. Sharabati | 1 | 14 | 2.26 |
Edward J. Wegman | 2 | 36 | 7.84 |
Yasmin H. Said | 3 | 14 | 1.92 |