A homophilic and dynamic influence maximization strategy based on independent cascade model in social networks
Influence maximization (IM) is crucial for recommendation systems and social networks. Previous research primarily focused on static networks, neglecting the homophily and dynamics inherent in real-world networks. This has led to inaccurate simulations of information spread and influence propagation...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphy.2024.1509905/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841561208980766720 |
|---|---|
| author | Gang Wang Shangyi Du Yurui Jiang Xianyong Li |
| author_facet | Gang Wang Shangyi Du Yurui Jiang Xianyong Li |
| author_sort | Gang Wang |
| collection | DOAJ |
| description | Influence maximization (IM) is crucial for recommendation systems and social networks. Previous research primarily focused on static networks, neglecting the homophily and dynamics inherent in real-world networks. This has led to inaccurate simulations of information spread and influence propagation between nodes, with traditional IM algorithms’ selected seed node sets failing to adapt to network evolution. To address this issue, this paper proposes a homophilic and dynamic influence maximization strategy based on independent cascade model (HDIM). Specifically, HDIM consists of two components: the seed node selection strategy that accounts for both homophily and dynamics (SSHD), and the independent cascade model based on influence homophily and dynamics (ICIHD). SSHD strictly constrains the proportions of different node types in the seed node set and can flexibly update the seed node set when the network structure changes. ICIHD redefines the propagation probabilities between nodes, adjusting them in response to changes in the network structure. Experimental results demonstrate HDIM’s excellent performance. Specifically, the influence range of HDIM exceeds that of state-of-the-art methods. Furthermore, the proportions of various activated nodes are closer to those in the original network. |
| format | Article |
| id | doaj-art-b20e6ab595dd440ab4f7e084b42696b9 |
| institution | Kabale University |
| issn | 2296-424X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physics |
| spelling | doaj-art-b20e6ab595dd440ab4f7e084b42696b92025-01-03T05:10:24ZengFrontiers Media S.A.Frontiers in Physics2296-424X2025-01-011210.3389/fphy.2024.15099051509905A homophilic and dynamic influence maximization strategy based on independent cascade model in social networksGang Wang0Shangyi Du1Yurui Jiang2Xianyong Li3School of Computer and Software Engineering, Xihua University, Chengdu, Sichuan, ChinaSchool of Statistics and Computer Science, McGill University, Montreal, QC, CanadaSchool of Computer and Software Engineering, Xihua University, Chengdu, Sichuan, ChinaSchool of Computer and Software Engineering, Xihua University, Chengdu, Sichuan, ChinaInfluence maximization (IM) is crucial for recommendation systems and social networks. Previous research primarily focused on static networks, neglecting the homophily and dynamics inherent in real-world networks. This has led to inaccurate simulations of information spread and influence propagation between nodes, with traditional IM algorithms’ selected seed node sets failing to adapt to network evolution. To address this issue, this paper proposes a homophilic and dynamic influence maximization strategy based on independent cascade model (HDIM). Specifically, HDIM consists of two components: the seed node selection strategy that accounts for both homophily and dynamics (SSHD), and the independent cascade model based on influence homophily and dynamics (ICIHD). SSHD strictly constrains the proportions of different node types in the seed node set and can flexibly update the seed node set when the network structure changes. ICIHD redefines the propagation probabilities between nodes, adjusting them in response to changes in the network structure. Experimental results demonstrate HDIM’s excellent performance. Specifically, the influence range of HDIM exceeds that of state-of-the-art methods. Furthermore, the proportions of various activated nodes are closer to those in the original network.https://www.frontiersin.org/articles/10.3389/fphy.2024.1509905/fullinfluence maximizationhomophilydynamicsindependent cascade modelsocial networks |
| spellingShingle | Gang Wang Shangyi Du Yurui Jiang Xianyong Li A homophilic and dynamic influence maximization strategy based on independent cascade model in social networks Frontiers in Physics influence maximization homophily dynamics independent cascade model social networks |
| title | A homophilic and dynamic influence maximization strategy based on independent cascade model in social networks |
| title_full | A homophilic and dynamic influence maximization strategy based on independent cascade model in social networks |
| title_fullStr | A homophilic and dynamic influence maximization strategy based on independent cascade model in social networks |
| title_full_unstemmed | A homophilic and dynamic influence maximization strategy based on independent cascade model in social networks |
| title_short | A homophilic and dynamic influence maximization strategy based on independent cascade model in social networks |
| title_sort | homophilic and dynamic influence maximization strategy based on independent cascade model in social networks |
| topic | influence maximization homophily dynamics independent cascade model social networks |
| url | https://www.frontiersin.org/articles/10.3389/fphy.2024.1509905/full |
| work_keys_str_mv | AT gangwang ahomophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks AT shangyidu ahomophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks AT yuruijiang ahomophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks AT xianyongli ahomophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks AT gangwang homophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks AT shangyidu homophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks AT yuruijiang homophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks AT xianyongli homophilicanddynamicinfluencemaximizationstrategybasedonindependentcascademodelinsocialnetworks |