Results 1 
2 of
2
Parameterized Approximability of Maximizing the Spread of Influence in Networks
"... Abstract. In this paper, we consider the problem of maximizing the spread of influence through a social network. Here, we are given a graph G = (V, E), a positive integer k and a threshold value thr(v) attached to each vertex v ∈ V. The objective is then to find a subset of k vertices to “activate ” ..."
Abstract

Cited by 6 (2 self)
 Add to MetaCart
Abstract. In this paper, we consider the problem of maximizing the spread of influence through a social network. Here, we are given a graph G = (V, E), a positive integer k and a threshold value thr(v) attached to each vertex v ∈ V. The objective is then to find a subset of k vertices to “activate ” such that the number of activated vertices at the end of a propagation process is maximum. A vertex v gets activated if at least thr(v) of its neighbors are. We show that this problem is strongly inapproximable in fpttime with respect to (w.r.t.) parameter k even for very restrictive thresholds. For unanimity thresholds, we prove that the problem is inapproximable in polynomial time and the decision version is W[1]hard w.r.t. parameter k. On the positive side, it becomes r(n)approximable in fpttime w.r.t. parameter k for any strictly increasing function r. Moreover, we give an fpttime algorithm to solve the decision version for bounded degree graphs. 1