Abstract
Heterogeneous information networks (HINs) are ubiquitous in real-world applications. Due to the heterogeneity in HINs, the typed edges may not fully align with each other. In order to capture the semantic subtlety, we propose the concept of aspects with each aspect being a unit representing one underlying semantic facet. Meanwhile, network embedding has emerged as a powerful method for learning network representation, where the learned embedding can be used as features in various downstream applications. Therefore, we are motivated to propose a novel embedding learning framework-ASPEM-to preserve the semantic information in HINs based on multiple aspects. Instead of preserving information of the network in one semantic space, ASPEM encapsulates information regarding each aspect individually. In order to select aspects for embedding purpose, we further devise a solution for ASPEM based on datasetwide statistics. To corroborate the efficacy of ASPEM, we conducted experiments on two real-words datasets with two types of applications-classification and link prediction. Experiment results demonstrate that ASPEM can outperform baseline network embedding learning methods by considering multiple aspects, where the aspects can be selected from the given HIN in an unsupervised manner.
| Original language | English (US) |
|---|---|
| Pages | 144-152 |
| Number of pages | 9 |
| State | Published - Jan 1 2018 |
| Event | 2018 SIAM International Conference on Data Mining, SDM 2018 - San Diego, United States Duration: May 3 2018 → May 5 2018 |
Other
| Other | 2018 SIAM International Conference on Data Mining, SDM 2018 |
|---|---|
| Country | United States |
| City | San Diego |
| Period | 5/3/18 → 5/5/18 |
Fingerprint
Keywords
- Graph mining
- Heterogeneous information networks
- Network embedding
- Representation learning
ASJC Scopus subject areas
- Computer Science Applications
- Software
Cite this
ASPEM : Embedding learning by aspects in heterogeneous information networks. / Shi, Yu; Gui, Huan; Zhu, Qi; Kaplan, Lance; Han, Jiawei.
2018. 144-152 Paper presented at 2018 SIAM International Conference on Data Mining, SDM 2018, San Diego, United States.Research output: Contribution to conference › Paper
}
TY - CONF
T1 - ASPEM
T2 - Embedding learning by aspects in heterogeneous information networks
AU - Shi, Yu
AU - Gui, Huan
AU - Zhu, Qi
AU - Kaplan, Lance
AU - Han, Jiawei
PY - 2018/1/1
Y1 - 2018/1/1
N2 - Heterogeneous information networks (HINs) are ubiquitous in real-world applications. Due to the heterogeneity in HINs, the typed edges may not fully align with each other. In order to capture the semantic subtlety, we propose the concept of aspects with each aspect being a unit representing one underlying semantic facet. Meanwhile, network embedding has emerged as a powerful method for learning network representation, where the learned embedding can be used as features in various downstream applications. Therefore, we are motivated to propose a novel embedding learning framework-ASPEM-to preserve the semantic information in HINs based on multiple aspects. Instead of preserving information of the network in one semantic space, ASPEM encapsulates information regarding each aspect individually. In order to select aspects for embedding purpose, we further devise a solution for ASPEM based on datasetwide statistics. To corroborate the efficacy of ASPEM, we conducted experiments on two real-words datasets with two types of applications-classification and link prediction. Experiment results demonstrate that ASPEM can outperform baseline network embedding learning methods by considering multiple aspects, where the aspects can be selected from the given HIN in an unsupervised manner.
AB - Heterogeneous information networks (HINs) are ubiquitous in real-world applications. Due to the heterogeneity in HINs, the typed edges may not fully align with each other. In order to capture the semantic subtlety, we propose the concept of aspects with each aspect being a unit representing one underlying semantic facet. Meanwhile, network embedding has emerged as a powerful method for learning network representation, where the learned embedding can be used as features in various downstream applications. Therefore, we are motivated to propose a novel embedding learning framework-ASPEM-to preserve the semantic information in HINs based on multiple aspects. Instead of preserving information of the network in one semantic space, ASPEM encapsulates information regarding each aspect individually. In order to select aspects for embedding purpose, we further devise a solution for ASPEM based on datasetwide statistics. To corroborate the efficacy of ASPEM, we conducted experiments on two real-words datasets with two types of applications-classification and link prediction. Experiment results demonstrate that ASPEM can outperform baseline network embedding learning methods by considering multiple aspects, where the aspects can be selected from the given HIN in an unsupervised manner.
KW - Graph mining
KW - Heterogeneous information networks
KW - Network embedding
KW - Representation learning
UR - http://www.scopus.com/inward/record.url?scp=85048335897&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85048335897&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:85048335897
SP - 144
EP - 152
ER -