Corvinus
Corvinus

Realistic models for diffusion of innovation

Sziklai, Balázs, Barnes, Kate and Pintér, József (2025) Realistic models for diffusion of innovation. Social Network Analysis and Mining, 15 (1). DOI 10.1007/s13278-025-01424-z

[img] PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
1MB

Official URL: https://doi.org/10.1007/s13278-025-01424-z


Abstract

It is widely acknowledged in the socio-economic literature that innovators and early adopters play an essential role in the diffusion of innovation. However, current approaches to influence maximization primarily concentrate on identifying influencers as the main targets of marketing campaigns. This perspective overlooks the fact that influencers and innovators seldom coincide, rendering the focus of influence maximization models misplaced. Practitioners often do not recognize that the influence maximization problem depends not only on the network structure but also on the underlying diffusion model. In this paper, we develop more realistic variants of the well-known linear threshold and independent cascade models. We incorporate the influence of adopter groups, which has a significant impact on the spread of influence. We leverage data from two different social networks and assess the efficiency of various centrality measures in the influence maximization model using both the traditional diffusion mechanisms and our novel approaches. The rankings obtained from the different models exhibit significant discrepancies implying that heuristics that perform well in a classical model may perform poorly in a more realistic setting. Instead of solely focusing on developing new algorithms for influence maximization, greater attention should be given to understanding and calibrating diffusion models to realistic settings.

Item Type:Article
Uncontrolled Keywords:Diffusion of innovation ; Innovators ; Early adopters ; Social networks ; Linear threshold ; Independent cascade
Divisions:Institute of Operations and Decision Sciences
Subjects:Decision making
Funders:János Bolyai Research Scholarship of the Hungarian Academy of Sciences, OTKA, New National Excellence Program, Artificial Intelligence National Laboratory, National Research, Development, and Innovation Office, TKP2021, Corvinus University of Budapest.
Projects:K 146320, ÚNKP-23-5, RRF-2.3.1-21-2022-00004, K142169, BME-NVA-02, Open access funding
DOI:10.1007/s13278-025-01424-z
ID Code:11037
Deposited By: MTMT SWORD
Deposited On:27 Mar 2025 14:07
Last Modified:27 Mar 2025 14:07

Repository Staff Only: item control page

Downloads

Downloads per month over past year

View more statistics