E.cient estimation of cumulative influence for multiple activation information diffusion model with continuous time delay

Authors:
Kazumi Saito;Masahiro Kimura;Kouzou Ohara;Hiroshi Motoda
Affiliations:
School of Administration and Informatics, University of Shizuoka, Shizuoka, Japan;Department of Electronics and Informatics, Ryukoku University, Otsu, Japan;Department of Integrated Information Technology, Aoyama Gakuin University, Kanagawa, Japan;Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan
Venue:
PRICAI'10 Proceedings of the 11th Pacific Rim international conference on Trends in artificial intelligence
Year:
2010

Citing 9
Cited 0

Maximizing the spread of influence through a social network

Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Information diffusion through blogspace

ACM SIGKDD Explorations Newsletter
The dynamics of viral marketing

EC '06 Proceedings of the 7th ACM conference on Electronic commerce
Blocking links to minimize contamination spread in a social network

ACM Transactions on Knowledge Discovery from Data (TKDD)
Extracting influential nodes for information diffusion on a social network

AAAI'07 Proceedings of the 22nd national conference on Artificial intelligence - Volume 2
Minimizing the spread of contamination by blocking links in a network

AAAI'08 Proceedings of the 23rd national conference on Artificial intelligence - Volume 2
Efficient estimation of influence functions for SIS model on social networks

IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
Discovering Influential Nodes for SIS Models in Social Networks

DS '09 Proceedings of the 12th International Conference on Discovery Science
Learning Continuous-Time Information Diffusion Model for Social Behavioral Data Analysis

ACML '09 Proceedings of the 1st Asian Conference on Machine Learning: Advances in Machine Learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

We show that the node cumulative influence for a particular class of information diffusion model in which a node can be activated multiple times, i.e. Susceptible/Infective/Susceptible (SIS) Model, can be very efficiently estimated in case of independent cascade (IC) framework with asynchronous time delay. The method exploits the property of continuous time delay within a stochastic framework and analytically derives the iterative formula to estimate cumulative influence without relying on awfully lengthy simulations. We show that it can accurately estimate the cumulative influence with much less computation time (about 2 to 6 orders of magnitude less) than the naive simulation using three real world social networks and thus it can be used to rank influential nodes quite effectively. Further, we show that the SIS model with a discrete time step, i.e. fixed synchronous time delay, gives adequate results only for a small time span.