Incremental algorithm for updating betweenness centrality in dynamically growing networks

Abstract

The increasing availability of dynamically growing digital data that can be used for extracting social networks has led to an upsurge of interest in the analysis of dynamic social networks. One key aspect of social network analysis is to understand the central nodes in a network. However, dynamic calculation of centrality values for rapidly growing networks might be unfeasibly expensive, especially if it involves recalculation from scratch for each time period. This paper proposes an incremental algorithm that effectively updates betweenness centralities of nodes in dynamic social networks while avoiding re-computations by exploiting information from earlier computations. Our performance results suggest that our incremental betweenness algorithm can achieve substantial performance speedup, on the order of thousands of times, over the state of the art, including the best-performing nonincremental betweenness algorithm and a recently proposed betweenness update algorithm.

Scalable Online Betweenness Centrality in Evolving Graphs

“…Lee et al [24] 2012 O(n 2 +m) 12k 65k Green et al [17] 2012 O(n 2 +nm) 23k 94k Kas et al [21] 2013 O(n 2 +nm) 8k 19k Nasre et al [28] 2014…”

“…Kas et al [21] extend the work proposed by Ramalingam and Reps [30] to accommodate the computation of vertex betweenness centrality while adding edges or vertices. Differently from Brandes' algorithm, their technique does not use dependencies but the actual shortest distances.…”

“…Therefore, we need to subtract the dependency on y. The subtracted value is adjusted by switching w with v in the dependency accumulation formula (lines [21][22]Alg. 4).…”

“…Table 3, the average performance of our framework is comparable to the reported results of previously proposed techniques. The method in [21] shows faster results in networks with low clustering coefficient where changes do not affect many vertices. However, in social networks with high clustering, the method by Kas et al [21] and especially QUBE [24] are slow due to high cost in updating vertex centrality (in QUBE, many vertices might be included in the minimum union cycle to be updated).…”

Betweenness Centrality – Incremental and Faster

“…Lee et al [14] present a framework called QUBE (Quick Update of BEtweenness centrality) which allows edges to be inserted and deleted from the graph, and recently, Singh et al [23] build on the work of Lee et al [14] to allow nodes to be added and deleted. Another recent work by Kas et al [11] deals with the problem of incremental BC and their algorithm is based on the dynamic all pairs shortest paths algorithm by Ramalingam and Reps [22]. All the above mentioned papers give encouraging experimental results, but there are no performance guarantees.…”