Machine learning to set hyperparameters for overlapping community detection algorithms
Abstract The Local Fitness Method (LFM) and Speaker‐Listener Label Propagation (SLPA) algorithms are widely used to detect overlapping communities in complex networks. The main problem with these two algorithms is that they are extremely sensitive to the setting of the hyperparameters. Previous meth...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-08-01
|
| Series: | The Journal of Engineering |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/tje2.12292 |
| _version_ | 1827860117383544832 |
|---|---|
| author | Chenglong Xiao Yajie Wang Shanshan Wang |
| author_facet | Chenglong Xiao Yajie Wang Shanshan Wang |
| author_sort | Chenglong Xiao |
| collection | DOAJ |
| description | Abstract The Local Fitness Method (LFM) and Speaker‐Listener Label Propagation (SLPA) algorithms are widely used to detect overlapping communities in complex networks. The main problem with these two algorithms is that they are extremely sensitive to the setting of the hyperparameters. Previous methods set the hyperparameters of LFM or SLPA based on either empirical values or random choices, resulting in a large amount of computation for adjusting those parameters. To solve this problem, in this paper, a machine‐learning‐based approach is proposed to automatically set the hyperparameters of these two algorithms. Experimental results show that compared with the manual method, automatically setting the hyperparameter using machine learning models can lead to higher‐quality divisions. Furthermore, in comparison to the well‐known hyperparameter tuning method using Bayesian Optimization, the proposed predictive model‐based approach can find suitable para meters for LFM and SLPA much faster, while achieving competitive results in terms of division quality. |
| first_indexed | 2024-03-12T13:18:42Z |
| format | Article |
| id | doaj.art-a1356f3f429943c982475844cd30f81c |
| institution | Directory Open Access Journal |
| issn | 2051-3305 |
| language | English |
| last_indexed | 2024-03-12T13:18:42Z |
| publishDate | 2023-08-01 |
| publisher | Wiley |
| record_format | Article |
| series | The Journal of Engineering |
| spelling | doaj.art-a1356f3f429943c982475844cd30f81c2023-08-26T10:20:49ZengWileyThe Journal of Engineering2051-33052023-08-0120238n/an/a10.1049/tje2.12292Machine learning to set hyperparameters for overlapping community detection algorithmsChenglong Xiao0Yajie Wang1Shanshan Wang2Department of Computer Science Shantou University Shantou ChinaDepartment of Computer Science Shantou University Shantou ChinaDepartment of Computer Science Shantou University Shantou ChinaAbstract The Local Fitness Method (LFM) and Speaker‐Listener Label Propagation (SLPA) algorithms are widely used to detect overlapping communities in complex networks. The main problem with these two algorithms is that they are extremely sensitive to the setting of the hyperparameters. Previous methods set the hyperparameters of LFM or SLPA based on either empirical values or random choices, resulting in a large amount of computation for adjusting those parameters. To solve this problem, in this paper, a machine‐learning‐based approach is proposed to automatically set the hyperparameters of these two algorithms. Experimental results show that compared with the manual method, automatically setting the hyperparameter using machine learning models can lead to higher‐quality divisions. Furthermore, in comparison to the well‐known hyperparameter tuning method using Bayesian Optimization, the proposed predictive model‐based approach can find suitable para meters for LFM and SLPA much faster, while achieving competitive results in terms of division quality.https://doi.org/10.1049/tje2.12292artificial intelligencecomplex networkscommunity detectionmachine learning |
| spellingShingle | Chenglong Xiao Yajie Wang Shanshan Wang Machine learning to set hyperparameters for overlapping community detection algorithms The Journal of Engineering artificial intelligence complex networks community detection machine learning |
| title | Machine learning to set hyperparameters for overlapping community detection algorithms |
| title_full | Machine learning to set hyperparameters for overlapping community detection algorithms |
| title_fullStr | Machine learning to set hyperparameters for overlapping community detection algorithms |
| title_full_unstemmed | Machine learning to set hyperparameters for overlapping community detection algorithms |
| title_short | Machine learning to set hyperparameters for overlapping community detection algorithms |
| title_sort | machine learning to set hyperparameters for overlapping community detection algorithms |
| topic | artificial intelligence complex networks community detection machine learning |
| url | https://doi.org/10.1049/tje2.12292 |
| work_keys_str_mv | AT chenglongxiao machinelearningtosethyperparametersforoverlappingcommunitydetectionalgorithms AT yajiewang machinelearningtosethyperparametersforoverlappingcommunitydetectionalgorithms AT shanshanwang machinelearningtosethyperparametersforoverlappingcommunitydetectionalgorithms |