A new deep sparse autoencoder for community detection in complex networks
Abstract Feature dimension reduction in the community detection is an important research topic in complex networks and has attracted many research efforts in recent years. However, most of existing algorithms developed for this purpose take advantage of classical mechanisms, which may be long experi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2020-05-01
|
| Series: | EURASIP Journal on Wireless Communications and Networking |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s13638-020-01706-4 |
| _version_ | 1828518707704365056 |
|---|---|
| author | Rong Fei Jingyuan Sha Qingzheng Xu Bo Hu Kan Wang Shasha Li |
| author_facet | Rong Fei Jingyuan Sha Qingzheng Xu Bo Hu Kan Wang Shasha Li |
| author_sort | Rong Fei |
| collection | DOAJ |
| description | Abstract Feature dimension reduction in the community detection is an important research topic in complex networks and has attracted many research efforts in recent years. However, most of existing algorithms developed for this purpose take advantage of classical mechanisms, which may be long experimental, time-consuming, and ineffective for complex networks. To this purpose, a novel deep sparse autoencoder for community detection, named DSACD, is proposed in this paper. In DSACD, a similarity matrix is constructed to reveal the indirect connections between nodes and a deep sparse automatic encoder based on unsupervised learning is designed to reduce the dimension and extract the feature structure of complex networks. During the process of back propagation, L-BFGS avoid the calculation of Hessian matrix which can increase the calculation speed. The performance of DSACD is validated on synthetic and real-world networks. Experimental results demonstrate the effectiveness of DSACD and the systematic comparisons with four algorithms confirm a significant improvement in terms of three index F same, NMI, and modularity Q. Finally, these achieved received signal strength indication (RSSI) data set can be aggregated into 64 correct communities, which further confirms its usability in indoor location systems. |
| first_indexed | 2024-12-11T18:59:18Z |
| format | Article |
| id | doaj.art-ed3d5a69e2d84d0e8732a88ccd13e8f7 |
| institution | Directory Open Access Journal |
| issn | 1687-1499 |
| language | English |
| last_indexed | 2024-12-11T18:59:18Z |
| publishDate | 2020-05-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EURASIP Journal on Wireless Communications and Networking |
| spelling | doaj.art-ed3d5a69e2d84d0e8732a88ccd13e8f72022-12-22T00:54:03ZengSpringerOpenEURASIP Journal on Wireless Communications and Networking1687-14992020-05-012020112510.1186/s13638-020-01706-4A new deep sparse autoencoder for community detection in complex networksRong Fei0Jingyuan Sha1Qingzheng Xu2Bo Hu3Kan Wang4Shasha Li5Xi’an University of TechnologyXi’an University of TechnologyCollege of Information and Communication, National University of DefenseBeijing Huadian Youkong Technology Co., Ltd.Xi’an University of TechnologyXi’an University of TechnologyAbstract Feature dimension reduction in the community detection is an important research topic in complex networks and has attracted many research efforts in recent years. However, most of existing algorithms developed for this purpose take advantage of classical mechanisms, which may be long experimental, time-consuming, and ineffective for complex networks. To this purpose, a novel deep sparse autoencoder for community detection, named DSACD, is proposed in this paper. In DSACD, a similarity matrix is constructed to reveal the indirect connections between nodes and a deep sparse automatic encoder based on unsupervised learning is designed to reduce the dimension and extract the feature structure of complex networks. During the process of back propagation, L-BFGS avoid the calculation of Hessian matrix which can increase the calculation speed. The performance of DSACD is validated on synthetic and real-world networks. Experimental results demonstrate the effectiveness of DSACD and the systematic comparisons with four algorithms confirm a significant improvement in terms of three index F same, NMI, and modularity Q. Finally, these achieved received signal strength indication (RSSI) data set can be aggregated into 64 correct communities, which further confirms its usability in indoor location systems.http://link.springer.com/article/10.1186/s13638-020-01706-4Community detectionDeep sparse autoencoderRSSI |
| spellingShingle | Rong Fei Jingyuan Sha Qingzheng Xu Bo Hu Kan Wang Shasha Li A new deep sparse autoencoder for community detection in complex networks EURASIP Journal on Wireless Communications and Networking Community detection Deep sparse autoencoder RSSI |
| title | A new deep sparse autoencoder for community detection in complex networks |
| title_full | A new deep sparse autoencoder for community detection in complex networks |
| title_fullStr | A new deep sparse autoencoder for community detection in complex networks |
| title_full_unstemmed | A new deep sparse autoencoder for community detection in complex networks |
| title_short | A new deep sparse autoencoder for community detection in complex networks |
| title_sort | new deep sparse autoencoder for community detection in complex networks |
| topic | Community detection Deep sparse autoencoder RSSI |
| url | http://link.springer.com/article/10.1186/s13638-020-01706-4 |
| work_keys_str_mv | AT rongfei anewdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT jingyuansha anewdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT qingzhengxu anewdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT bohu anewdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT kanwang anewdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT shashali anewdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT rongfei newdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT jingyuansha newdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT qingzhengxu newdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT bohu newdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT kanwang newdeepsparseautoencoderforcommunitydetectionincomplexnetworks AT shashali newdeepsparseautoencoderforcommunitydetectionincomplexnetworks |