A Probability PPV Model for Social Network Influence Maximization Problem

DOI

10.2991/aiie-16.2016.9How to use a DOI?

Keywords

influence maximization; viral marketing; PageRank algorithm; personalization

Abstract

For Influence Maximization(IM) problem based on social network, effective and personalized probability learning method was still not theoretical guaranteed. In this paper, we proposed a PPV probability model based on IM problem, which effectively learnt influence probabilities and personal-ized influence for each node pair. By clustering user groups and analyzing similarity of users from both offline action log and online social network topological structure, we estimated reliable parameters of our probability model, differed user's influence with different features on its neighbors, and improved probability learning procedure. PageRank algorithm and fuzzy cognitive map concept help to validate our model in PPV calculation. Experiments show that our approach outperforms the state-of-the-art algorithm. As preference property r and transition property p, which is called PPV in our model, explains features of influence between user pairs, accuracy of personalized influence probability is improved.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yunjia Ge
AU  - Dong Wang
PY  - 2016/11
DA  - 2016/11
TI  - A Probability PPV Model for Social Network Influence Maximization Problem
BT  - Proceedings of the 2016 2nd International Conference on Artificial Intelligence and Industrial Engineering (AIIE 2016)
PB  - Atlantis Press
SP  - 37
EP  - 42
SN  - 1951-6851
UR  - https://doi.org/10.2991/aiie-16.2016.9
DO  - 10.2991/aiie-16.2016.9
ID  - Ge2016/11
ER  -