### Citations

1511 |
Community structure in social and biological networks
- Girvan, ME
- 2002
(Show Context)
Citation Context ...ade model by removing edges greedily, which we show does not yield approximations with guarantees. They compare their method to two other heuristics based on outdegree and edge betweenness centrality =-=[8]-=-, which we will use as baselines in our study as well. Tong et. al. [20] also consider removing edges from the network but under a different cascade model, for which the eigenvalue of the adjacency ma... |

750 |
An analysis of approximation for maximizing submodular set functions
- Nemhauser, Wolsey, et al.
- 1978
(Show Context)
Citation Context ...ing to the chosen set the next best edge to remove in terms of marginal decrease in susceptability, until k edges have been selected, produces solutions that are within (1 − 1/e) of the optimal value =-=[17, 14]-=-. We also prove that other network manipulation operations, such as adding edges, deleting nodes and adding nodes, are supermodular, as described in the Appendix 5. We conduct computational experiment... |

690 |
Threshold models of collective behavior.
- Granovetter
- 1978
(Show Context)
Citation Context ...es to the network topology would result in suppressing the influence of an undesireable diffusion process, in the best possible way? First, we consider the linear threshold model as a diffusion model =-=[10]-=-. This model is widely adopted by sociologists as representative of adoption dynamics, where each node or individual has a threshold, representing the fraction of its 1054 055 056 057 058 059 060 061... |

568 | Mining the network value of customers
- Domingos, Richardson
- 2001
(Show Context)
Citation Context ...ta show that our method significantly outperforms other common heuristics. 1 Introduction The diffusion of ideas and influence is a topic of study across many disciplines ranging from viral marketing =-=[6]-=- and population epidemics [11], to social media [15] and other fields. A central element in any diffusion process is the communication channel along which the spread occurs. Recent efforts in network ... |

548 | A faster algorithm for betweenness centrality
- Brandes
- 2001
(Show Context)
Citation Context ...he k edges with highest edge betweenness centrality, where this measure is defined for edge e as the sum of the fraction of all-pairs shortest paths that pass through e (referred to as ’Betweenness’) =-=[3]-=-, (4) select the k edges whose destination nodes have the highest out-degree (referred to as ’Degree’), (5) select the k edges with highest diffusion probability (weight) wu,v, where an edge goes from... |

490 | The mathematics of infectious diseases
- Hethcote
(Show Context)
Citation Context ...ficantly outperforms other common heuristics. 1 Introduction The diffusion of ideas and influence is a topic of study across many disciplines ranging from viral marketing [6] and population epidemics =-=[11]-=-, to social media [15] and other fields. A central element in any diffusion process is the communication channel along which the spread occurs. Recent efforts in network analysis, as well as the proli... |

363 | Meme-tracking and the dynamics of the news cycle.
- Leskovec, Backstrom, et al.
- 2009
(Show Context)
Citation Context ...ther common heuristics. 1 Introduction The diffusion of ideas and influence is a topic of study across many disciplines ranging from viral marketing [6] and population epidemics [11], to social media =-=[15]-=- and other fields. A central element in any diffusion process is the communication channel along which the spread occurs. Recent efforts in network analysis, as well as the proliferation of real-world... |

322 | Costeffective outbreak detection in networks
- Leskovec, Guestrin, et al.
- 2007
(Show Context)
Citation Context ...le of live-edge graphs. Using the greedy approach in a naive way will result in evaluating marginal gain for each candidate edge at every iteration. Instead, we use a technique called lazy evaluation =-=[16]-=-, which avoids computing the function for all edges and has been shown to result in significant speed-ups over the naive evaluation. 7378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 3... |

183 | Scalable influence maximization for prevalent viral marketing in large-scale social networks.
- Chen, Wang, et al.
- 2010
(Show Context)
Citation Context ...ld networks, and compare the quality of the solution it provides against other heuristic algorithms. Since evaluating the true expected susceptability of a graph has been shown to be #P -hard problem =-=[4]-=-, we use the usual Monte Carlo based approach and approximate it by the average susceptability over a large sample of live-edge graphs. Using the greedy approach in a naive way will result in evaluati... |

166 | Epidemic spreading in real networks: An eigenvalue viewpoint
- Wang, Chakrabarti, et al.
- 2003
(Show Context)
Citation Context ...ect k edges uniformly at random (referred to as ’Random’), (2) select the k edges that cause the maximum decrease in the leading eigenvalue of the network when removed from it (referred to as Eigen’) =-=[21, 20]-=-, (3) select the k edges with highest edge betweenness centrality, where this measure is defined for edge e as the sum of the fraction of all-pairs shortest paths that pass through e (referred to as ’... |

160 |
Éva Tardos. Maximizing the spread of influence through a social network
- Kempe, Kleinberg
- 2003
(Show Context)
Citation Context ...n this representation, a number of methods have been devised with the goal of finding a set of k nodes whose adoption of a given idea would result in maximizing the spread of this idea in the network =-=[6, 12]-=-. This particular problem has been extended for different diffusion models [18], and increasingly efficient solutions are being proposed for it [7]. However, not much attention has been accorded to th... |

75 | Scalable influence maximization in social networks under the linear threshold model.
- Chen, Yuan, et al.
- 2010
(Show Context)
Citation Context ...ich the eigenvalue of the adjacency matrix determines the epidemic threshold. While the Independent Cascade model has been well studied, fewer have considered the Linear Threshold model. Chen et. al. =-=[5]-=- study influence maximization under the Linear Threshold Model and show that computing exact influence in general networks is #P − hard. They propose a more scalable method for estimating influence un... |

63 | A: Inferring networks of diffusion and influence
- MG, Leskovec, et al.
- 2010
(Show Context)
Citation Context ... news media site or blog, and each edge e(u, v) represents the recorded event of v copying u. These edges are inferred from actual hyperlink cascade traces using a network inference algorithm, NETINF =-=[9]-=-. To assign probabilities on the edges, we make use of the median transmission time, also provided as part of the dataset. Let ˜tu,v be the median transmission time between two nodes u and v, then we ... |

56 | Uncovering the Temporal Dynamics of Diffusion Networks.
- Gomez-Rodriguez, Balduzzi, et al.
- 2011
(Show Context)
Citation Context ...ing a set of k nodes whose adoption of a given idea would result in maximizing the spread of this idea in the network [6, 12]. This particular problem has been extended for different diffusion models =-=[18]-=-, and increasingly efficient solutions are being proposed for it [7]. However, not much attention has been accorded to the study of negative phenomena that propagate in networks. Diseases that spread ... |

24 | Maximizing the spread of cascades using network design.
- Sheldon, Dilkina, et al.
- 2012
(Show Context)
Citation Context ... Experiments show that our method significantly outperforms the other heuristics. Related Work: The topic of manipulating network structure to impact diffusion processes has been recently explored in =-=[20, 19, 13, 2]-=-. Several studies consider manipulating nodes. Sheldon et. al. [19] solve the problem of adding nodes to the network to maximize spread under the Independent Cascade Model using Sample Average Approxi... |

21 |
Submodular function maximization. Tractability: Practical Approaches to Hard Problems,
- Krause, Golovin
- 2012
(Show Context)
Citation Context ...ing to the chosen set the next best edge to remove in terms of marginal decrease in susceptability, until k edges have been selected, produces solutions that are within (1 − 1/e) of the optimal value =-=[17, 14]-=-. We also prove that other network manipulation operations, such as adding edges, deleting nodes and adding nodes, are supermodular, as described in the Appendix 5. We conduct computational experiment... |

9 |
Aditya Prakash, Tina Eliassi-Rad, Michalis Faloutsos, and Christos Faloutsos. 2012. Gelling, and melting, large graphs by edge manipulation
- Tong, B
(Show Context)
Citation Context ... Experiments show that our method significantly outperforms the other heuristics. Related Work: The topic of manipulating network structure to impact diffusion processes has been recently explored in =-=[20, 19, 13, 2]-=-. Several studies consider manipulating nodes. Sheldon et. al. [19] solve the problem of adding nodes to the network to maximize spread under the Independent Cascade Model using Sample Average Approxi... |

5 |
Scalable influence estimation in continuous time diffusion networks
- Du, Song, et al.
- 2013
(Show Context)
Citation Context ...aximizing the spread of this idea in the network [6, 12]. This particular problem has been extended for different diffusion models [18], and increasingly efficient solutions are being proposed for it =-=[7]-=-. However, not much attention has been accorded to the study of negative phenomena that propagate in networks. Diseases that spread via contagion in societies, rumors that diffuse through blogs and ne... |

3 |
Robust protection of networks against cascading phenomena,
- Bogunovic
- 2012
(Show Context)
Citation Context ... Experiments show that our method significantly outperforms the other heuristics. Related Work: The topic of manipulating network structure to impact diffusion processes has been recently explored in =-=[20, 19, 13, 2]-=-. Several studies consider manipulating nodes. Sheldon et. al. [19] solve the problem of adding nodes to the network to maximize spread under the Independent Cascade Model using Sample Average Approxi... |

1 |
Hiroshi Motoda. Blocking links to minimize contamination spread in a social network
- Kimura, Saito
(Show Context)
Citation Context |