Dual Quaternion Embeddings for Link Prediction
The applications of knowledge graph have received much attention in the field of artificial intelligence. The quality of knowledge graphs is, however, often influenced by missing facts. To predict the missing facts, various solid transformation based models have been proposed by mapping knowledge gr...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/11/12/5572 |
id |
doaj-61750127b4f04adfb80f7c409083080f |
---|---|
record_format |
Article |
spelling |
doaj-61750127b4f04adfb80f7c409083080f2021-07-01T00:20:40ZengMDPI AGApplied Sciences2076-34172021-06-01115572557210.3390/app11125572Dual Quaternion Embeddings for Link PredictionLiming Gao0Huiling Zhu1Hankz Hankui Zhuo2Jin Xu3School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510000, ChinaCollege of Information Science and Technology, Jinan University, Guangzhou 510000, ChinaSchool of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510000, ChinaData Quality Team, WeChat, Tencent Inc., Guangzhou 510000, ChinaThe applications of knowledge graph have received much attention in the field of artificial intelligence. The quality of knowledge graphs is, however, often influenced by missing facts. To predict the missing facts, various solid transformation based models have been proposed by mapping knowledge graphs into low dimensional spaces. However, most of the existing transformation based approaches ignore that there are multiple relations between two entities, which is common in the real world. In order to address this challenge, we propose a novel approach called DualQuatE that maps entities and relations into a dual quaternion space. Specifically, entities are represented by pure quaternions and relations are modeled based on the combination of rotation and translation from head to tail entities. After that we utilize interactions of different translations and rotations to distinguish various relations between head and tail entities. Experimental results exhibit that the performance of DualQuatE is competitive compared to the existing state-of-the-art models.https://www.mdpi.com/2076-3417/11/12/5572knowledge graph embeddinglink predictionartificial intelligence |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Liming Gao Huiling Zhu Hankz Hankui Zhuo Jin Xu |
spellingShingle |
Liming Gao Huiling Zhu Hankz Hankui Zhuo Jin Xu Dual Quaternion Embeddings for Link Prediction Applied Sciences knowledge graph embedding link prediction artificial intelligence |
author_facet |
Liming Gao Huiling Zhu Hankz Hankui Zhuo Jin Xu |
author_sort |
Liming Gao |
title |
Dual Quaternion Embeddings for Link Prediction |
title_short |
Dual Quaternion Embeddings for Link Prediction |
title_full |
Dual Quaternion Embeddings for Link Prediction |
title_fullStr |
Dual Quaternion Embeddings for Link Prediction |
title_full_unstemmed |
Dual Quaternion Embeddings for Link Prediction |
title_sort |
dual quaternion embeddings for link prediction |
publisher |
MDPI AG |
series |
Applied Sciences |
issn |
2076-3417 |
publishDate |
2021-06-01 |
description |
The applications of knowledge graph have received much attention in the field of artificial intelligence. The quality of knowledge graphs is, however, often influenced by missing facts. To predict the missing facts, various solid transformation based models have been proposed by mapping knowledge graphs into low dimensional spaces. However, most of the existing transformation based approaches ignore that there are multiple relations between two entities, which is common in the real world. In order to address this challenge, we propose a novel approach called DualQuatE that maps entities and relations into a dual quaternion space. Specifically, entities are represented by pure quaternions and relations are modeled based on the combination of rotation and translation from head to tail entities. After that we utilize interactions of different translations and rotations to distinguish various relations between head and tail entities. Experimental results exhibit that the performance of DualQuatE is competitive compared to the existing state-of-the-art models. |
topic |
knowledge graph embedding link prediction artificial intelligence |
url |
https://www.mdpi.com/2076-3417/11/12/5572 |
work_keys_str_mv |
AT liminggao dualquaternionembeddingsforlinkprediction AT huilingzhu dualquaternionembeddingsforlinkprediction AT hankzhankuizhuo dualquaternionembeddingsforlinkprediction AT jinxu dualquaternionembeddingsforlinkprediction |
_version_ |
1721348853024686080 |