TY - GEN
T1 - HISBmodel
T2 - 25th International Conference on Neural Information Processing, ICONIP 2018
AU - Hosni, Adil Imad Eddine
AU - Li, Kan
AU - Ahmed, Sadique
N1 - Publisher Copyright:
© 2018, Springer Nature Switzerland AG.
PY - 2018
Y1 - 2018
N2 - This paper attempts to address the rumor propagation problem in online social networks (OSNs) and proposes a novel rumor diffusion model, named the HISBmodel. Its originality lies in the consideration of various human factors such as the human social and individual behaviors and the individuals’ opinions. Moreover, we present new metrics that allow accurate assessment of the propagation of rumors. Based on this model, we present a strategy to minimize the influence of the rumor. Instead of blocking nodes, we propose to launch a truth campaign to raise the awareness to prevent the influence of a rumor. This problem is formulated from the perspective of a network inference using the survival theory. The experimental results illustrate that the HISBmodel depicts the evolution of rumor propagation more realistic than classical models. Moreover, Our model highlights the impact of human factors accurately as proven in the studies of the literature. Finally, these experiments showed the outstanding performance of our strategy to minimize the influence of the rumor by selecting precisely the candidate nodes to diminish the influence of the rumor.
AB - This paper attempts to address the rumor propagation problem in online social networks (OSNs) and proposes a novel rumor diffusion model, named the HISBmodel. Its originality lies in the consideration of various human factors such as the human social and individual behaviors and the individuals’ opinions. Moreover, we present new metrics that allow accurate assessment of the propagation of rumors. Based on this model, we present a strategy to minimize the influence of the rumor. Instead of blocking nodes, we propose to launch a truth campaign to raise the awareness to prevent the influence of a rumor. This problem is formulated from the perspective of a network inference using the survival theory. The experimental results illustrate that the HISBmodel depicts the evolution of rumor propagation more realistic than classical models. Moreover, Our model highlights the impact of human factors accurately as proven in the studies of the literature. Finally, these experiments showed the outstanding performance of our strategy to minimize the influence of the rumor by selecting precisely the candidate nodes to diminish the influence of the rumor.
KW - Humans individual behaviors
KW - Humans social behaviors
KW - Rumor influence minimization
KW - Rumor propagation
UR - http://www.scopus.com/inward/record.url?scp=85059068505&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-04179-3_2
DO - 10.1007/978-3-030-04179-3_2
M3 - Conference contribution
AN - SCOPUS:85059068505
SN - 9783030041786
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 14
EP - 27
BT - Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings
A2 - Leung, Andrew Chi Sing
A2 - Ozawa, Seiichi
A2 - Cheng, Long
PB - Springer Verlag
Y2 - 13 December 2018 through 16 December 2018
ER -