Abstract
Attributed network embedding enables joint representation learning of node links and attributes. Existing attributed network embedding models are designed in continuous Euclidean spaces which often introduce data redundancy and impose challenges to storage and computation costs. To this end, we present a Binarized Attributed Network Embedding model (BANE for short) to learn binary node representation. Specifically, we define a new Weisfeiler-Lehman proximity matrix to capture data dependence between node links and attributes by aggregating the information of node attributes and links from neighboring nodes to a given target node in a layer-wise manner. Based on the Weisfeiler-Lehman proximity matrix, we formulate a new Weisfiler-Lehman matrix factorization learning function under the binary node representation constraint. The learning problem is a mixed integer optimization and an efficient cyclic coordinate descent (CCD) algorithm is used as the solution. Node classification and link prediction experiments on real-world datasets show that the proposed BANE model outperforms the state-of-the-art network embedding methods.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDM 2018 |
| Subtitle of host publication | 17–20 November 2018 Singapore |
| Editors | Dacheng Tao, Bhavani Thuraisingham |
| Place of Publication | Piscataway NJ USA |
| Publisher | IEEE, Institute of Electrical and Electronics Engineers |
| Pages | 1476-1481 |
| Number of pages | 6 |
| ISBN (Electronic) | 9781538691595, 9781538691588 |
| ISBN (Print) | 9781538691601 |
| DOIs | |
| Publication status | Published - 2018 |
| Externally published | Yes |
| Event | IEEE International Conference on Data Mining 2018 - Singapore, Singapore Duration: 17 Nov 2018 → 20 Nov 2018 http://icdm2018.org/ |
Conference
| Conference | IEEE International Conference on Data Mining 2018 |
|---|---|
| Abbreviated title | ICDM 2018 |
| Country | Singapore |
| City | Singapore |
| Period | 17/11/18 → 20/11/18 |
| Internet address |
Keywords
- Attributed network embedding
- Learning to hash
- Weisfeiler-Lehman graph kernels
Cite this
}
Binarized attributed network embedding. / Yang, Hong; Pan, Shirui; Zhang, Peng; Chen, Ling; Lian, Defu; Zhang, Chengqi.
Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDM 2018: 17–20 November 2018 Singapore. ed. / Dacheng Tao; Bhavani Thuraisingham. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2018. p. 1476-1481 8626170.Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review
TY - GEN
T1 - Binarized attributed network embedding
AU - Yang, Hong
AU - Pan, Shirui
AU - Zhang, Peng
AU - Chen, Ling
AU - Lian, Defu
AU - Zhang, Chengqi
PY - 2018
Y1 - 2018
N2 - Attributed network embedding enables joint representation learning of node links and attributes. Existing attributed network embedding models are designed in continuous Euclidean spaces which often introduce data redundancy and impose challenges to storage and computation costs. To this end, we present a Binarized Attributed Network Embedding model (BANE for short) to learn binary node representation. Specifically, we define a new Weisfeiler-Lehman proximity matrix to capture data dependence between node links and attributes by aggregating the information of node attributes and links from neighboring nodes to a given target node in a layer-wise manner. Based on the Weisfeiler-Lehman proximity matrix, we formulate a new Weisfiler-Lehman matrix factorization learning function under the binary node representation constraint. The learning problem is a mixed integer optimization and an efficient cyclic coordinate descent (CCD) algorithm is used as the solution. Node classification and link prediction experiments on real-world datasets show that the proposed BANE model outperforms the state-of-the-art network embedding methods.
AB - Attributed network embedding enables joint representation learning of node links and attributes. Existing attributed network embedding models are designed in continuous Euclidean spaces which often introduce data redundancy and impose challenges to storage and computation costs. To this end, we present a Binarized Attributed Network Embedding model (BANE for short) to learn binary node representation. Specifically, we define a new Weisfeiler-Lehman proximity matrix to capture data dependence between node links and attributes by aggregating the information of node attributes and links from neighboring nodes to a given target node in a layer-wise manner. Based on the Weisfeiler-Lehman proximity matrix, we formulate a new Weisfiler-Lehman matrix factorization learning function under the binary node representation constraint. The learning problem is a mixed integer optimization and an efficient cyclic coordinate descent (CCD) algorithm is used as the solution. Node classification and link prediction experiments on real-world datasets show that the proposed BANE model outperforms the state-of-the-art network embedding methods.
KW - Attributed network embedding
KW - Learning to hash
KW - Weisfeiler-Lehman graph kernels
UR - http://www.scopus.com/inward/record.url?scp=85061386533&partnerID=8YFLogxK
U2 - 10.1109/ICDM.2018.8626170
DO - 10.1109/ICDM.2018.8626170
M3 - Conference Paper
SN - 9781538691601
SP - 1476
EP - 1481
BT - Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDM 2018
A2 - Tao, Dacheng
A2 - Thuraisingham, Bhavani
PB - IEEE, Institute of Electrical and Electronics Engineers
CY - Piscataway NJ USA
ER -