CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling
Heterogeneous information network (HIN) embedding is an important tool for tasks such as node classification, community detection, and recommendation. It aims to find the representations of nodes that preserve the proximity between entities of different nature. A family of approaches that are widely...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/24/2/276 |
| _version_ | 1797480572376842240 |
|---|---|
| author | Ling Zhan Tao Jia |
| author_facet | Ling Zhan Tao Jia |
| author_sort | Ling Zhan |
| collection | DOAJ |
| description | Heterogeneous information network (HIN) embedding is an important tool for tasks such as node classification, community detection, and recommendation. It aims to find the representations of nodes that preserve the proximity between entities of different nature. A family of approaches that are widely adopted applies random walk to generate a sequence of heterogeneous contexts, from which, the embedding is learned. However, due to the multipartite graph structure of HIN, hub nodes tend to be over-represented to their context in the sampled sequence, giving rise to imbalanced samples of the network. Here, we propose a new embedding method: CoarSAS2hvec. The self-avoiding short sequence sampling with the HIN coarsening procedure (CoarSAS) is utilized to better collect the rich information in HIN. An optimized loss function is used to improve the performance of the HIN structure embedding. CoarSAS2hvec outperforms nine other methods in node classification and community detection on four real-world data sets. Using entropy as a measure of the amount of information, we confirm that CoarSAS catches richer information of the network compared with that through other methods. Hence, the traditional loss function applied to samples by CoarSAS can also yield improved results. Our work addresses a limitation of the random-walk-based HIN embedding that has not been emphasized before, which can shed light on a range of problems in HIN analyses. |
| first_indexed | 2024-03-09T22:01:59Z |
| format | Article |
| id | doaj.art-d328d94264ac44b0927b6a95da54efa1 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-09T22:01:59Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-d328d94264ac44b0927b6a95da54efa12023-11-23T19:48:47ZengMDPI AGEntropy1099-43002022-02-0124227610.3390/e24020276CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network SamplingLing Zhan0Tao Jia1College of Computer and Information Science, Southwest University, Chongqing 400715, ChinaCollege of Computer and Information Science, Southwest University, Chongqing 400715, ChinaHeterogeneous information network (HIN) embedding is an important tool for tasks such as node classification, community detection, and recommendation. It aims to find the representations of nodes that preserve the proximity between entities of different nature. A family of approaches that are widely adopted applies random walk to generate a sequence of heterogeneous contexts, from which, the embedding is learned. However, due to the multipartite graph structure of HIN, hub nodes tend to be over-represented to their context in the sampled sequence, giving rise to imbalanced samples of the network. Here, we propose a new embedding method: CoarSAS2hvec. The self-avoiding short sequence sampling with the HIN coarsening procedure (CoarSAS) is utilized to better collect the rich information in HIN. An optimized loss function is used to improve the performance of the HIN structure embedding. CoarSAS2hvec outperforms nine other methods in node classification and community detection on four real-world data sets. Using entropy as a measure of the amount of information, we confirm that CoarSAS catches richer information of the network compared with that through other methods. Hence, the traditional loss function applied to samples by CoarSAS can also yield improved results. Our work addresses a limitation of the random-walk-based HIN embedding that has not been emphasized before, which can shed light on a range of problems in HIN analyses.https://www.mdpi.com/1099-4300/24/2/276heterogeneous information networksnetwork embeddingcontext samplingrandom walkinformation entropy |
| spellingShingle | Ling Zhan Tao Jia CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling Entropy heterogeneous information networks network embedding context sampling random walk information entropy |
| title | CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling |
| title_full | CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling |
| title_fullStr | CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling |
| title_full_unstemmed | CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling |
| title_short | CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling |
| title_sort | coarsas2hvec heterogeneous information network embedding with balanced network sampling |
| topic | heterogeneous information networks network embedding context sampling random walk information entropy |
| url | https://www.mdpi.com/1099-4300/24/2/276 |
| work_keys_str_mv | AT lingzhan coarsas2hvecheterogeneousinformationnetworkembeddingwithbalancednetworksampling AT taojia coarsas2hvecheterogeneousinformationnetworkembeddingwithbalancednetworksampling |