DOI QR코드

DOI QR Code

소셜 네트워크를 위한 확산기반 영향력 극대화 기법

Diffusion-Based Influence Maximization Method for Social Network

  • Nguyen, Tri-Hai (Soongsil University, School of Electronic Engineering) ;
  • Yoo, Myungsik (Soongsil University, School of Electronic Engineering)
  • 투고 : 2016.10.11
  • 심사 : 2016.10.14
  • 발행 : 2016.10.31

초록

정보 확산 극대화 문제는 소셜 네트워크에서 정보 확산을 최대로 할 수 있는 Seed 노드 군을 설정하는 것이다. 기존의 Greedy 알고리즘은 최적에 근접한 해를 제시하였으나 높은 계산량의 문제가 있다. 몇몇 Heuristic 알고리즘들이 계산량 감소를 목표로 제안되었으나 정보 확산 성능 측면에서 한계점이 있다. 본 논문에서는 General Degree Discount 알고리즘을 제안하고, 제안된 알고리즘이 계산량 측면 및 정보 확산 성능 측면에서 기존 Heuristic 알고리즘 대비 우수한 성능을 보임을 입증하고자 한다.

Influence maximization problem is to select seed node set, which maximizes information spread in social networks. Greedy algorithm shows an optimum solution, but has a high computational cost. A few heuristic algorithms were proposed to reduce the complexity, but their performance in influence maximization is limited. In this paper, we propose general degree discount algorithm, and show that it has better performance while keeping complexity low.

키워드

참고문헌

  1. Y. Kim and G. Park, "Topic sensitive_social relation rank algorithm for efficient social search," J. KICS, vol. 38, no. 5, pp. 385-393, 2013.
  2. D. Kempe, J. Kleinberg, and E. Tardos, "Maximizing the spread of influence through a social network," in Proc. 9th ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, pp. 137-146, 2003.
  3. M. Kimura and K. Saito, "Tractable models for information diffusion in social networks," in Knowledge Discovery in Databases: PKDD, Springer Berlin Heidelberg, pp. 259-271, 2006.
  4. W. Chen, Y. Wang, and S. Siyu, "Efficient influence maximization in social networks," in Proc. 15th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pp. 199-208, 2009.
  5. J. Leskovec, Stanford Large Network Dataset Collection - Social networks, Available: https://snap.stanford.edu/data/