Empirical Analysis of Various Seed Selection Methods

Rabadiya, Kinjal; Patel, Ritesh

doi:10.1007/978-981-15-7062-9_39

Cited by 1 publication

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The higher degree nodes take less time to reach more nodes. [38]. 5) Degree Discount [39] : A degree discount heuristic algorithm which provides much better results than the classical degree and centrality based heuristic algorithms.. 6) CELF [10] : Cost Effective Lazy Forward (CELF) is focus on contaminant detection for water distribution network, finding important stories in a blog network and 700 times faster than the greedy algorithm.…”

Section: A Influence Maximizationmentioning

confidence: 99%

Seed Selection Algorithm using Centrality based Entropy

Rabadiya¹,

Patel²

2023

IJCDS

View full text Add to dashboard Cite

Recently, Business expansion, marketing and advertisement is more fast and convenient process through social network analytic. In this paper, the influence maximization problem is addressed, which is the process of selecting the best suitable initial users or customers or spreaders who can use or advertise or spread the product information in such way that in their own social network maximum people can receive the information about that product. Still, the seed selection problem is NP-hard problem and to date none of the algorithm has focus on combination of various centrality of nodes that can significantly impact on seed selection process. In this paper, we propose the novel seed selection algorithm which can fill the gap and achieve the diffusion speed by combining five centrality of node. For that, We conduct simulations to evaluate the diffusion speed of our proposed algorithm and existing benchmark seed selection algorithms using real-world authors collaboration networks. Experimental results show that our proposed algorithm outperforms various existing benchmark seed selection algorithms by achieving optimal diffusion speed.

show abstract

Section: A Influence Maximizationmentioning

confidence: 99%