Probabilistic Coarsening for Knowledge Graph Embeddings
Knowledge graphs have risen in popularity in recent years, demonstrating their utility in applications across the spectrum of computer science. Finding their embedded representations is thus highly desirable as it makes them easily operated on and reasoned with by machines. With this in mind, we pro...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Axioms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1680/12/3/275 |
| _version_ | 1797613456815292416 |
|---|---|
| author | Marcin Pietrasik Marek Z. Reformat |
| author_facet | Marcin Pietrasik Marek Z. Reformat |
| author_sort | Marcin Pietrasik |
| collection | DOAJ |
| description | Knowledge graphs have risen in popularity in recent years, demonstrating their utility in applications across the spectrum of computer science. Finding their embedded representations is thus highly desirable as it makes them easily operated on and reasoned with by machines. With this in mind, we propose a simple meta-strategy for embedding knowledge graphs using probabilistic coarsening. In this approach, a knowledge graph is first coarsened before being embedded by an arbitrary embedding method. The resulting coarse embeddings are then extended down as those of the initial knowledge graph. Although straightforward, this allows for faster training by reducing knowledge graph complexity while revealing its higher-order structures. We demonstrate this empirically on four real-world datasets, which show that coarse embeddings are learned faster and are often of higher quality. We conclude that coarsening is a recommended prepossessing step regardless of the underlying embedding method used. |
| first_indexed | 2024-03-11T06:56:08Z |
| format | Article |
| id | doaj.art-0e67fd17fd6e4cc2a21f702804cb0e4f |
| institution | Directory Open Access Journal |
| issn | 2075-1680 |
| language | English |
| last_indexed | 2024-03-11T06:56:08Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Axioms |
| spelling | doaj.art-0e67fd17fd6e4cc2a21f702804cb0e4f2023-11-17T09:35:15ZengMDPI AGAxioms2075-16802023-03-0112327510.3390/axioms12030275Probabilistic Coarsening for Knowledge Graph EmbeddingsMarcin Pietrasik0Marek Z. Reformat1Department of Electrical and Computer Engineering, University of Alberta, Edmonton, AB T6G 2R3, CanadaDepartment of Electrical and Computer Engineering, University of Alberta, Edmonton, AB T6G 2R3, CanadaKnowledge graphs have risen in popularity in recent years, demonstrating their utility in applications across the spectrum of computer science. Finding their embedded representations is thus highly desirable as it makes them easily operated on and reasoned with by machines. With this in mind, we propose a simple meta-strategy for embedding knowledge graphs using probabilistic coarsening. In this approach, a knowledge graph is first coarsened before being embedded by an arbitrary embedding method. The resulting coarse embeddings are then extended down as those of the initial knowledge graph. Although straightforward, this allows for faster training by reducing knowledge graph complexity while revealing its higher-order structures. We demonstrate this empirically on four real-world datasets, which show that coarse embeddings are learned faster and are often of higher quality. We conclude that coarsening is a recommended prepossessing step regardless of the underlying embedding method used.https://www.mdpi.com/2075-1680/12/3/275knowledge graphembeddingcoarsening |
| spellingShingle | Marcin Pietrasik Marek Z. Reformat Probabilistic Coarsening for Knowledge Graph Embeddings Axioms knowledge graph embedding coarsening |
| title | Probabilistic Coarsening for Knowledge Graph Embeddings |
| title_full | Probabilistic Coarsening for Knowledge Graph Embeddings |
| title_fullStr | Probabilistic Coarsening for Knowledge Graph Embeddings |
| title_full_unstemmed | Probabilistic Coarsening for Knowledge Graph Embeddings |
| title_short | Probabilistic Coarsening for Knowledge Graph Embeddings |
| title_sort | probabilistic coarsening for knowledge graph embeddings |
| topic | knowledge graph embedding coarsening |
| url | https://www.mdpi.com/2075-1680/12/3/275 |
| work_keys_str_mv | AT marcinpietrasik probabilisticcoarseningforknowledgegraphembeddings AT marekzreformat probabilisticcoarseningforknowledgegraphembeddings |