文章目录
1.综述
- embedding 方法
- OpenEA: “A Benchmarking Study of Embedding-based Entity Alignment for Knowledge Graphs”.
Zequn Sun, Qingheng Zhang, Wei Hu, Chengming Wang, Muhao Chen, Farahnaz Akrami, Chengkai Li. PVLDB, vol. 13. ACM 2020 [paper][code][笔记] - "An Experimental Study of State-of-the-Art Entity Alignment Approaches".
Xiang Zhao, Weixin Zeng, Jiuyang Tang, Wei Wang, Fabian Suchanek. TKDE, 2020 [paper][笔记]
2.技术论文
-
JE: “A Joint Embedding Method for Entity Alignment of Knowledge Bases”.
Yanchao Hao, Yuanzhe Zhang, Shizhu He, Kang Liu, Jun Zhao. (CCKS 2016) [paper][code] -
MTransE: “Multilingual Knowledge Graph Embeddings for Cross-lingual Knowledge Alignment”.
Muhao Chen, Yingtao Tian, Mohan Yang, Carlo Zaniolo. (IJCAI 2017) [paper][code] -
JAPE: “Cross-Lingual Entity Alignment via Joint Attribute-Preserving Embedding”.
Zequn Sun, Wei Hu, Chengkai Li. (ISWC 2017) [paper][code] -
IPTransE: “Iterative Entity Alignment via Joint Knowledge Embeddings”.
Hao Zhu, Ruobing Xie, Zhiyuan Liu, Maosong Sun. (IJCAI 2017) [paper][code] -
BootEA: “Bootstrapping Entity Alignment with Knowledge Graph Embedding”.
Zequn Sun, Wei Hu, Qingheng Zhang, Yuzhong Qu. (IJCAI 2018) [paper][code][笔记] -
KDCoE: “Co-training Embeddings of Knowledge Graphs and Entity Descriptions for Cross-lingual Entity Alignment”.
Muhao Chen, Yingtao Tian, Kai-Wei Chang, Steven Skiena, Carlo Zaniolo. (IJCAI 2018) [paper][code] -
NTAM: “Non-translational Alignment for Multi-relational Networks”.
Shengnan Li, Xin Li, Rui Ye, Mingzhong Wang, Haiping Su, Yingzi Ou. (IJCAI 2018) [paper][code] -
**“LinkNBed: Multi-Graph Representation Learning with Entity Linkage”.""
Rakshit Trivedi, Bunyamin Sisman, Jun Ma, Christos Faloutsos, Hongyuan Zha, Xin Luna Dong (ACL 2018) [paper] -
GCN-Align: “Cross-lingual Knowledge Graph Alignment via Graph Convolutional Networks”.
Zhichun Wang, Qingsong Lv, Xiaohan Lan, Yu Zhang. (EMNLP 2018) [paper][code] -
AttrE: “Entity Alignment between Knowledge Graphs Using Attribute Embeddings”.
Bayu D. Trsedya, Jianzhong Qi, Rui Zhang. (AAAI 2019) [paper][code] -
SEA: “Semi-Supervised Entity Alignment via Knowledge Graph Embedding with Awareness of Degree Difference”.
Shichao Pei, Lu Yu, Robert Hoehndorf, Xiangliang Zhang. (WWW 2019) [paper][code] -
RSN4EA: “Learning to Exploit Long-term Relational Dependencies in Knowledge Graphs”.
Lingbing Guo, Zequn Sun, Wei Hu. (ICML 2019) [paper][code] -
MuGNN: “Multi-Channel Graph Neural Network for Entity Alignment”.
Yixin Cao, Zhiyuan Liu, Chengjiang Li, Zhiyuan Liu, Juanzi Li, Tat-Seng Chua. (ACL 2019) [paper][code] -
GMNN: “Cross-lingual Knowledge Graph Alignment via Graph Matching Neural Network”.
Kun Xu, Liwei Wang, Mo Yu, Yansong Feng, Yan Song, Zhiguo Wang, Dong Yu. (ACL 2019) [paper][code] -
MultiKE: “Multi-view Knowledge Graph Embedding for Entity Alignment”.
Qingheng Zhang, Zequn Sun, Wei Hu, Muhao Chen, Lingbing Guo, Yuzhong Qu. (IJCAI 2019) [paper][code] -
RDGCN: “Relation-Aware Entity Alignment for Heterogeneous Knowledge Graphs”.
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Rui Yan, Dongyan Zhao. (IJCAI 2019) [paper][code] -
OTEA: “Improving Cross-lingual Entity Alignment via Optimal Transport”.
Shichao Pei, Lu Yu, Xiangliang Zhang. (IJCAI 2019) [paper][code] -
NAEA: “Neighborhood-Aware Attentional Representation for Multilingual Knowledge Graphs”.
Qiannan Zhu, Xiaofei Zhou, Jia Wu, Jianlong Tan, Li Guo. (IJCAI 2019) [paper][code] -
AVR-GCN: “A Vectorized Relational Graph Convolutional Network for Multi-Relational Network Alignment”.
Rui Ye, Xin Li, Yujie Fang, Hongyu Zang, Mingzhong Wang. (IJCAI 2019) [paper][code] -
TransEdge: “TransEdge: Translating Relation-Contextualized Embeddings for Knowledge Graphs”.
Zequn Sun, Jiacheng Huang, Wei Hu, Muhao Chen, Lingbing Guo, Yuzhong Qu. (ISWC 2019) [paper][code] -
KECG: “Semi-supervised Entity Alignment via Joint Knowledge Embedding Model and Cross-graph Model”.
Chengjiang Li, Yixin Cao, Lei Hou, Jiaxin Shi, Juanzi Li, Tat-Seng Chua. (EMNLP 2019) [paper][code] -
HGCN: “Jointly Learning Entity and Relation Representations for Entity Alignment”.
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Dongyan Zhao. (EMNLP 2019) [paper][code] -
MMEA: “Modeling Multi-mapping relations for Precise Cross-lingual Entity Alignment”.
Xiaofei Shi, Yanghua Xiao. (EMNLP 2019) [paper][code] -
HMAN: “Aligning Cross-lingual Entities with Multi-Aspect Information”.
Hsiu-Wei Yang, Yanyan Zou, Peng Shi, Wei Lu, Jimmy Lin, Xu Sun. (EMNLP 2019) [paper][code] -
AKE: “Guiding Cross-lingual Entity Alignment via Adversarial Knowledge Embedding”.
Xixun Lin, Hong Yang, Jia Wu, Chuan Zhou, Bin Wang. (ICDM 2019) [paper][code] -
MRAEA: “MRAEA: An Efficient and Robust Cross-lingual Entity Alignment Approach via Meta Relation Aware Representation”.
Xin Mao, Wenting Wang, Huimin Xu, Man Lan, Yuanbin Wu. (WSDM 2020) [paper][code] -
AliNet: “Knowledge Graph Alignment Network with Gated Multi-hop Neighborhood Aggregation”.
Zequn Sun, Chengming Wang, Wei Hu, Muhao Chen, Jian Dai, Wei Zhang, Yuzhong Qu. (AAAI 2020) [paper][code] -
"Coordinated Reasoning for Cross-Lingual Knowledge Graph Alignment".
Kun Xu, Linfeng Song, Yansong Feng, Yan Song, Dong Yu. (AAAI 2020) [paper][code] -
COTSAE: “COTSAE: CO-Training of Structure and Attribute Embeddings for Entity Alignment”.
Kai Yang, Shaoqin Liu, Junfeng Zhao, Yasha Wang, Bing Xie. (AAAI 2020) [paper][code] -
CEAFF: “Collective Embedding-based Entity Alignment via Adaptive Features”.
Weixin Zeng, Xiang Zhao, Jiuyang Tang, Xuemin Lin. (ICDE 2020) [paper][code] -
"Deep Graph Matching Consensus".
Matthias Fey, Jan E. Lenssen, Christopher Morris, Jonathan Masci, Nils M. Kriege. (ICLR 2020) [paper][code] -
CG-MuAlign: “Collective Multi-type Entity Alignment Between Knowledge Graphs”.
Qi Zhu, Hao Wei, Bunyamin Sisman, Da Zheng, Christos Faloutsos, Xin Luna Dong, Jiawei Han. (WWW 2020) [paper][code] -
JarKA: “JarKA: Modeling Attribute Interactions for Cross-lingual Knowledge Alignment”.
Bo Chen, Jing Zhang, Xiaobin Tang, Hong Chen, Cuiping Li. (PAKDD 2020) [paper][code] -
NMN: “Neighborhood Matching Network for Entity Alignment”.
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Dongyan Zhao. (ACL 2020) [paper][code] -
BERT-INT: “BERT-INT: A BERT-based Interaction Model For Knowledge Graph Alignment”.
Xiaobin Tang, Jing Zhang, Bo Chen, Yang Yang, Hong Chen, Cuiping Li. (IJCAI 2020) [paper][code] -
SSP: “Global Structure and Local Semantics-Preserved Embeddings for Entity Alignment”.
Hao Nie, Xianpei Han, Le Sun, Chi Man Wong, Qiang Chen, Suhui Wu, Wei Zhang. (IJCAI 2020) [paper][code] -
DAT: “Degree-Aware Alignment for Entities in Tail”.
Weixin Zeng, Xiang Zhao, Wei Wang, Jiuyang Tang, Zhen Tan. (SIGIR 2020) [paper][code] -
RREA: “Relational Reflection Entity Alignment”.
Xin Mao, Wenting Wang, Huimin Xu, Yuanbin Wu, Man Lan. (CIKM 2020) [paper][code] -
REA: “REA: Robust Cross-lingual Entity Alignment Between Knowledge Graphs”.
Shichao Pei, Lu Yu, Guoxian Yu, Xiangliang Zhang. (KDD 2020) [paper][code] -
HyperKA: “Knowledge Association with Hyperbolic Knowledge Graph Embeddings”.
Zequn Sun, Muhao Chen, Wei Hu, Chengming Wang, Jian Dai, Wei Zhang. (EMNLP 2020) [paper][code] -
AttrGNN: “Exploring and Evaluating Attributes, Values, and Structures for Entity Alignment”.
Zhiyuan Liu, Yixin Cao, Liangming Pan, Juanzi Li, Zhiyuan Liu, Tat-Seng Chua. (EMNLP 2020) [paper][code] -
EPEA: “Knowledge Graph Alignment with Entity-Pair Embedding”.
Zhichun Wang, Jinjian Yang, Xiaoju Ye. (EMNLP 2020) [paper] -
"Learning Short-Term Differences and Long-Term Dependencies for Entity Alignment".
Jia Chen, Zhixu Li, Pengpeng Zhao, An Liu, Lei Zhao, Zhigang Chen, Xiangliang Zhang. (ISWC 2020) [paper] -
"Visual Pivoting for (Unsupervised) Entity Alignment".
Fangyu Liu, Muhao Chen, Dan Roth, Nigel Collier. (AAAI 2021) [paper][code] -
DINGAL: “Dynamic Knowledge Graph Alignment”.
Yuchen Yan, Lihui Liu, Yikun Ban, Baoyu Jing, Hanghang Tong. (AAAI 2021) [paper] -
RNM: “Relation-Aware Neighborhood Matching Model for Entity Alignment”.
Yao Zhu, Hongzhi Liu, Zhonghai Wu, Yingpeng Du. (AAAI 2021) [paper] -
"Cross-lingual Entity Alignment with Incidental Supervision".
Muhao Chen, Weijia Shi, Ben Zhou, Dan Roth. (EACL 2021) [paper][code] -
"Active Learning for Entity Alignment".
Max Berrendorf, Evgeniy Faerman, Volker Tresp. (ECIR 2021) [paper] -
Dual-AMN: “Boosting the Speed of Entity Alignment 10×: Dual Attention Matching Network with Normalized Hard Sample Mining”.
Xin Mao, Wenting Wang, Yuanbin Wu, Man Lan. (WWW 2021) [paper][code]1. JE: “A Joint Embedding Method for Entity Alignment of Knowledge Bases”.
Yanchao Hao, Yuanzhe Zhang, Shizhu He, Kang Liu, Jun Zhao. (CCKS 2016) [paper][code] -
MTransE: “Multilingual Knowledge Graph Embeddings for Cross-lingual Knowledge Alignment”.
Muhao Chen, Yingtao Tian, Mohan Yang, Carlo Zaniolo. (IJCAI 2017) [paper][code] -
JAPE: “Cross-Lingual Entity Alignment via Joint Attribute-Preserving Embedding”.
Zequn Sun, Wei Hu, Chengkai Li. (ISWC 2017) [paper][code] -
IPTransE: “Iterative Entity Alignment via Joint Knowledge Embeddings”.
Hao Zhu, Ruobing Xie, Zhiyuan Liu, Maosong Sun. (IJCAI 2017) [paper][code] -
BootEA: “Bootstrapping Entity Alignment with Knowledge Graph Embedding”.
Zequn Sun, Wei Hu, Qingheng Zhang, Yuzhong Qu. (IJCAI 2018) [paper][code] -
KDCoE: “Co-training Embeddings of Knowledge Graphs and Entity Descriptions for Cross-lingual Entity Alignment”.
Muhao Chen, Yingtao Tian, Kai-Wei Chang, Steven Skiena, Carlo Zaniolo. (IJCAI 2018) [paper][code] -
NTAM: “Non-translational Alignment for Multi-relational Networks”.
Shengnan Li, Xin Li, Rui Ye, Mingzhong Wang, Haiping Su, Yingzi Ou. (IJCAI 2018) [paper][code] -
**“LinkNBed: Multi-Graph Representation Learning with Entity Linkage”.""
Rakshit Trivedi, Bunyamin Sisman, Jun Ma, Christos Faloutsos, Hongyuan Zha, Xin Luna Dong (ACL 2018) [paper] -
GCN-Align: “Cross-lingual Knowledge Graph Alignment via Graph Convolutional Networks”.
Zhichun Wang, Qingsong Lv, Xiaohan Lan, Yu Zhang. (EMNLP 2018) [paper][code] -
AttrE: “Entity Alignment between Knowledge Graphs Using Attribute Embeddings”.
Bayu D. Trsedya, Jianzhong Qi, Rui Zhang. (AAAI 2019) [paper][code] -
SEA: “Semi-Supervised Entity Alignment via Knowledge Graph Embedding with Awareness of Degree Difference”.
Shichao Pei, Lu Yu, Robert Hoehndorf, Xiangliang Zhang. (WWW 2019) [paper][code] -
RSN4EA: “Learning to Exploit Long-term Relational Dependencies in Knowledge Graphs”.
Lingbing Guo, Zequn Sun, Wei Hu. (ICML 2019) [paper][code] -
MuGNN: “Multi-Channel Graph Neural Network for Entity Alignment”.
Yixin Cao, Zhiyuan Liu, Chengjiang Li, Zhiyuan Liu, Juanzi Li, Tat-Seng Chua. (ACL 2019) [paper][code] -
GMNN: “Cross-lingual Knowledge Graph Alignment via Graph Matching Neural Network”.
Kun Xu, Liwei Wang, Mo Yu, Yansong Feng, Yan Song, Zhiguo Wang, Dong Yu. (ACL 2019) [paper][code] -
MultiKE: “Multi-view Knowledge Graph Embedding for Entity Alignment”.
Qingheng Zhang, Zequn Sun, Wei Hu, Muhao Chen, Lingbing Guo, Yuzhong Qu. (IJCAI 2019) [paper][code] -
RDGCN: “Relation-Aware Entity Alignment for Heterogeneous Knowledge Graphs”.
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Rui Yan, Dongyan Zhao. (IJCAI 2019) [paper][code] -
OTEA: “Improving Cross-lingual Entity Alignment via Optimal Transport”.
Shichao Pei, Lu Yu, Xiangliang Zhang. (IJCAI 2019) [paper][code] -
NAEA: “Neighborhood-Aware Attentional Representation for Multilingual Knowledge Graphs”.
Qiannan Zhu, Xiaofei Zhou, Jia Wu, Jianlong Tan, Li Guo. (IJCAI 2019) [paper][code] -
AVR-GCN: “A Vectorized Relational Graph Convolutional Network for Multi-Relational Network Alignment”.
Rui Ye, Xin Li, Yujie Fang, Hongyu Zang, Mingzhong Wang. (IJCAI 2019) [paper][code] -
TransEdge: “TransEdge: Translating Relation-Contextualized Embeddings for Knowledge Graphs”.
Zequn Sun, Jiacheng Huang, Wei Hu, Muhao Chen, Lingbing Guo, Yuzhong Qu. (ISWC 2019) [paper][code] -
KECG: “Semi-supervised Entity Alignment via Joint Knowledge Embedding Model and Cross-graph Model”.
Chengjiang Li, Yixin Cao, Lei Hou, Jiaxin Shi, Juanzi Li, Tat-Seng Chua. (EMNLP 2019) [paper][code] -
HGCN: “Jointly Learning Entity and Relation Representations for Entity Alignment”.
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Dongyan Zhao. (EMNLP 2019) [paper][code] -
MMEA: “Modeling Multi-mapping relations for Precise Cross-lingual Entity Alignment”.
Xiaofei Shi, Yanghua Xiao. (EMNLP 2019) [paper][code] -
HMAN: “Aligning Cross-lingual Entities with Multi-Aspect Information”.
Hsiu-Wei Yang, Yanyan Zou, Peng Shi, Wei Lu, Jimmy Lin, Xu Sun. (EMNLP 2019) [paper][code] -
AKE: “Guiding Cross-lingual Entity Alignment via Adversarial Knowledge Embedding”.
Xixun Lin, Hong Yang, Jia Wu, Chuan Zhou, Bin Wang. (ICDM 2019) [paper][code] -
MRAEA: “MRAEA: An Efficient and Robust Cross-lingual Entity Alignment Approach via Meta Relation Aware Representation”.
Xin Mao, Wenting Wang, Huimin Xu, Man Lan, Yuanbin Wu. (WSDM 2020) [paper][code] -
AliNet: “Knowledge Graph Alignment Network with Gated Multi-hop Neighborhood Aggregation”.
Zequn Sun, Chengming Wang, Wei Hu, Muhao Chen, Jian Dai, Wei Zhang, Yuzhong Qu. (AAAI 2020) [paper][code] -
"Coordinated Reasoning for Cross-Lingual Knowledge Graph Alignment".
Kun Xu, Linfeng Song, Yansong Feng, Yan Song, Dong Yu. (AAAI 2020) [paper][code] -
COTSAE: “COTSAE: CO-Training of Structure and Attribute Embeddings for Entity Alignment”.
Kai Yang, Shaoqin Liu, Junfeng Zhao, Yasha Wang, Bing Xie. (AAAI 2020) [paper][code] -
CEAFF: “Collective Embedding-based Entity Alignment via Adaptive Features”.
Weixin Zeng, Xiang Zhao, Jiuyang Tang, Xuemin Lin. (ICDE 2020) [paper][code] -
"Deep Graph Matching Consensus".
Matthias Fey, Jan E. Lenssen, Christopher Morris, Jonathan Masci, Nils M. Kriege. (ICLR 2020) [paper][code] -
CG-MuAlign: “Collective Multi-type Entity Alignment Between Knowledge Graphs”.
Qi Zhu, Hao Wei, Bunyamin Sisman, Da Zheng, Christos Faloutsos, Xin Luna Dong, Jiawei Han. (WWW 2020) [paper][code] -
JarKA: “JarKA: Modeling Attribute Interactions for Cross-lingual Knowledge Alignment”.
Bo Chen, Jing Zhang, Xiaobin Tang, Hong Chen, Cuiping Li. (PAKDD 2020) [paper][code] -
NMN: “Neighborhood Matching Network for Entity Alignment”.
Yuting Wu, Xiao Liu, Yansong Feng, Zheng Wang, Dongyan Zhao. (ACL 2020) [paper][code] -
BERT-INT: “BERT-INT: A BERT-based Interaction Model For Knowledge Graph Alignment”.
Xiaobin Tang, Jing Zhang, Bo Chen, Yang Yang, Hong Chen, Cuiping Li. (IJCAI 2020) [paper][code] -
SSP: “Global Structure and Local Semantics-Preserved Embeddings for Entity Alignment”.
Hao Nie, Xianpei Han, Le Sun, Chi Man Wong, Qiang Chen, Suhui Wu, Wei Zhang. (IJCAI 2020) [paper][code] -
DAT: “Degree-Aware Alignment for Entities in Tail”.
Weixin Zeng, Xiang Zhao, Wei Wang, Jiuyang Tang, Zhen Tan. (SIGIR 2020) [paper][code] -
RREA: “Relational Reflection Entity Alignment”.
Xin Mao, Wenting Wang, Huimin Xu, Yuanbin Wu, Man Lan. (CIKM 2020) [paper][code] -
REA: “REA: Robust Cross-lingual Entity Alignment Between Knowledge Graphs”.
Shichao Pei, Lu Yu, Guoxian Yu, Xiangliang Zhang. (KDD 2020) [paper][code] -
HyperKA: “Knowledge Association with Hyperbolic Knowledge Graph Embeddings”.
Zequn Sun, Muhao Chen, Wei Hu, Chengming Wang, Jian Dai, Wei Zhang. (EMNLP 2020) [paper][code] -
AttrGNN: “Exploring and Evaluating Attributes, Values, and Structures for Entity Alignment”.
Zhiyuan Liu, Yixin Cao, Liangming Pan, Juanzi Li, Zhiyuan Liu, Tat-Seng Chua. (EMNLP 2020) [paper][code] -
EPEA: “Knowledge Graph Alignment with Entity-Pair Embedding”.
Zhichun Wang, Jinjian Yang, Xiaoju Ye. (EMNLP 2020) [paper] -
"Learning Short-Term Differences and Long-Term Dependencies for Entity Alignment".
Jia Chen, Zhixu Li, Pengpeng Zhao, An Liu, Lei Zhao, Zhigang Chen, Xiangliang Zhang. (ISWC 2020) [paper] -
"Visual Pivoting for (Unsupervised) Entity Alignment".
Fangyu Liu, Muhao Chen, Dan Roth, Nigel Collier. (AAAI 2021) [paper][code] -
DINGAL: “Dynamic Knowledge Graph Alignment”.
Yuchen Yan, Lihui Liu, Yikun Ban, Baoyu Jing, Hanghang Tong. (AAAI 2021) [paper] -
RNM: “Relation-Aware Neighborhood Matching Model for Entity Alignment”.
Yao Zhu, Hongzhi Liu, Zhonghai Wu, Yingpeng Du. (AAAI 2021) [paper] -
"Cross-lingual Entity Alignment with Incidental Supervision".
Muhao Chen, Weijia Shi, Ben Zhou, Dan Roth. (EACL 2021) [paper][code] -
"Active Learning for Entity Alignment".
Max Berrendorf, Evgeniy Faerman, Volker Tresp. (ECIR 2021) [paper] -
Dual-AMN: “Boosting the Speed of Entity Alignment 10×: Dual Attention Matching Network with Normalized Hard Sample Mining”.
Xin Mao, Wenting Wang, Yuanbin Wu, Man Lan. (WWW 2021) [paper][code]
3.汇总
3.1定义
- 匹配两个KG或一个KG内指向同一物理对象,合并向同时提
定义统一
- Entity Linking=entity disambiguation
- Entity resolution=entity matching=deduplication=record linkage
EA
EA
-
分类:
- Scope:
- entity alignment<-本文只考虑这个
- relation
- 类别对齐:class of taxonomies of two KGs
- 方法:有一次性执行三种任务的joint model
- Background knowledge
- OAEI:使用ontology(T-box)作为背景信息
- 另一种:不使用ontology的方法
- Training
- 无监督:PARIS,SIGMa,AML
- 有监督:基于pre-defined mappings的
- 半监督:bootstrapping(self-training,co-training)
- Scope:
-
EA with deep leaning:
- 基于graph representation learning technologies
- 建模KG结构
- 生成实体嵌入
- 基于graph representation learning technologies
-
比较
- 无监督
- PARIS
- Agreement-MakerLight(AML):使用背景信息(本体)
- ER方法:基于名称的启发式方法
- goal相同:EA=ER–因为相同所以比较ER方法
- 无监督
-
Bechmarks:
- 语言内+DBPedia
- DBP15K
- DWY15
- 问题:现有的Bechmarks,只包含schema和instance信息。对不假设有可用的本体的EA方法来说。–所以本文不介绍本体?
- 语言内+DBPedia
-
PS:
- OAEI:推广了KG track
- 不公平
3.2 评价指标
- 对齐质量:准确性和全面性
- MR
- MRR
- Hits@m:m=1为precision
- precision/recall/f1
- 传统方法再用
- 对齐效率:分区索引技术对候选匹配对的筛选能力和准确性
- 缩减率
- 候选对完整性
- 候选对质量
3.3 数据集
-
Embedding数据集
- FBK15
- FBK15-237
- WN18
- WN18RR
-
传统实体对齐数据集:
- OAEI(since 2004)
-
embedding实体对齐数据集
-
DBP15K:
- 跨语言:
- zh-en,
- zh:关系三元组数:70414,关系数1701,属性三元组数:248035
- en: 关系三元组数:95142,关系数1323,属性三元组数:343218
- ja-en,
- ja:关系三元组数:77214,关系数1299,属性三元组数:248991
- en: 关系三元组数:93484,关系数1153,属性三元组数:320616
- fr-en
- fr:关系三元组数:105998,关系数903,属性三元组数:273825
- en: 关系三元组数:115722,关系数1208,属性三元组数:351094
- zh-en,
- 实体对齐连接数:15k(每对语言间)
- 度的分布:大多在1,从2-10,度越大,实体数量下降
- DBPedia
- 跨语言:
-
WK3L
-
DWY100K:
- 每个KG实体数:100k
- 单语言:
- DBP-WD,
- DBP:关系三元组数:463294,关系数330,属性三元组数:341770
- WD:关系三元组数:448774,关系数220,属性三元组数:779402
- DBP-YG
- DBP:关系三元组数:428952,关系数302,属性三元组数:383757
- YG:关系三元组数:502563,关系数31,属性三元组数:98028
- (DBP:DBPedia,YG:Yago3,WD:wikidata)
- DBP-WD,
- 每对有100k个实体对齐连接
- 度的分布:没有度为1or2的,峰值在4,之后递减
-
SRPRS
- 认为以前的数据集太稠密了(DBP,DWY),度的分布偏离现实
- 跨语言:
- EN-FR,
- EN:关系三元组数:36508,关系数221,属性三元组数:60800
- FR:关系三元组数:33532,关系数177,属性三元组数:53045
- EN-DE
- EN:关系三元组数:38363,关系数220,属性三元组数:55580
- DE:关系三元组数:37377,关系数120,属性三元组数:73753
- EN-FR,
- 单语言:
- DBP-WD,
- DBP:关系三元组数:33421,关系数253,属性三元组数:64021
- WD:关系三元组数:40159,关系数144,属性三元组数:133371
- DBP-YG
- DBP:关系三元组数:33748,关系数223,属性三元组数:58853
- YG:关系三元组数:36569,关系数30,属性三元组数:18241
- DBP-WD,
- 每种有15k个实体对齐连接
- 度的分布:很现实
- 度小的实体多(精心取样)
-
DBP-FB(An Experimental Study of State-of-the-Art Entity Alignment Approaches)
- DBP: 关系三元组数:96414,关系数407,属性三元组数:127614
- FB:关系三元组数:111974,关系数882,属性三元组数:78740
-
-
度的分布
-
EN-FR的统计
3.4 数据预处理技术
3.5 索引
- 分区索引:过滤掉不可能匹配的实体对,降低计算复杂度,避免数据库规模二次增长
3.6 对齐
3.6.1 按属性相似度/文本相似度做:成对实体对齐
- 传统概率模型:
- 基于属性相似度评分–>三分类:匹配,可能匹配,不匹配
- 也可用01
- 机器学习的模型
- 根据实体属性构建向量
- 方法:决策树、SVM等分类模型
- 优点:自动拟合属性间的组合关系和对应程度,减少人为介入
- 可引入无监督、半监督
- 文本匹配/语义匹配
- 文本特征明显的实体匹配
- 实体简介很长的那种
- Bert什么的
3.6.2 协同对齐:考虑不同实体间的关联
在属性相似度基础上考虑了结构相似度
3.6.2.1 局部实体对齐
- 计算相似度
- 考虑邻居的属性(带匹配实体对的邻居属性集合)
- 但不把邻居节点当做平等的实体去计算结构相似性
- 计算
- s i m ( e i , e j ) = α ⋅ s i m a t t r ( e i , e j ) + ( 1 − α ) ⋅ s i m N B ( e i , e j ) 实 体 本 身 的 相 似 度 : s i m a t t r ( e i , e j ) = Σ ( a 1 , a 2 ) ∈ A t t r ( e i , e j ) s i m ( a 1 , a 2 ) 实 体 关 联 实 体 相 似 度 s i m N B ( e i , e j ) = Σ ( e i ′ , e j ′ ) ∈ N B ( e i , e j ) s i m a t t r ( e i ′ , e j ′ ) sim(e_i,e_j)=\alpha \cdot sim_{attr}(e_i,e_j)+(1-\alpha)\cdot sim_{NB}(e_i,e_j)\\ 实体本身的相似度:sim_{attr}(e_i,e_j)=\Sigma_{(a_1,a_2)\in Attr(e_i,e_j)}sim(a_1,a_2)\\ 实体关联实体相似度sim_{NB}(e_i,e_j)=\Sigma_{(e_i',e_j')\in NB(e_i,e_j) sim_{attr}(e_i',e_j')} sim(ei,ej)=α⋅simattr(ei,ej)+(1−α)⋅simNB(ei,ej)实体本身的相似度:simattr(ei,ej)=Σ(a1,a2)∈Attr(ei,ej)sim(a1,a2)实体关联实体相似度simNB(ei,ej)=Σ(ei′,ej′)∈NB(ei,ej)simattr(ei′,ej′)
3.6.2.2 全局实体对齐
- 通过不同匹配策略之间相互影响调整实体之间的相似度
- 基于相似度传播的方法
- 基本思想:通过seed alignment以bootstrapping的方式迭代的产生一些新的匹配
- 半监督?
- 基于概率模型的方法
- 基本思想:全局概率最大化。通过为实体匹配关系和匹配决策决策复杂的概率模型,来避免bootstrapping–需要人工参与
- 基本方法:贝叶斯网络/LDA/CRF/Markov
3.6.3 基于embedding的方法分类
A Benchmarking Study of Embedding-based Entity Alignment for Knowledge Graphs
<-分类来源
-
Embedding Module
- 关系嵌入
- triple embedding:transE
- Path:路径上的长期依赖
- Neighbor:GCN
- 属性嵌入
- 属性
- literal
- 关系嵌入
-
Interaction Mode
- Combination mode
- transformation:学习映射M
- embedding space Calibration:嵌入到同一空间
- parameter sharing:同一向量表示
- parameter swapping: ( e 1 , e 2 ) ∈ S ′ , 则 ( e 1 , r 1 , e 1 ′ ) − > ( e 2 , r 1 , e 1 ′ ) (e_1,e_2)\in S',则(e_1,r_1,e_1')->(e_2,r_1,e_1') (e1,e2)∈S′,则(e1,r1,e1′)−>(e2,r1,e1′)
- learning
- 监督
- 半监督
- 无监督
- Combination mode
-
Embedding
- transE
- GCN
-
Alignment
- 2个向量映射到一个空间
- 训练一个相同的向量
- Transition
- Corpus-fusion
- Margin-based
- Graph matching
- Attribution refined
-
Prediction:
- 相似度计算:
- cosine
- euclidean
- Manhattan distance
- 相似度计算:
-
Extra information Module
- 用以增强EA
- 方法
- bootstrapping(or self-learning:
- 利用置信度高的对齐结果加入训练数据(下个iteration)
- multi-type literal information
- 属性
- 实体描述
- 实体名
- 完善KG的结构
- bootstrapping(or self-learning:
-
模块级别的比较
- 在个模块下介绍各方法如何实现该模块
4.开源代码
- OpenEA
- 开源的embedding pipeline组件库
5.效果比较
6.使用场景
- 单语言/多语言
- 稀疏/稠密
- 大规模/中等规模
- 1v1/多对多
- 1v1:BootEA
7. 实验效果
7.1 DBP15k
- DBP15k
- DBP15k
- 组1:仅用结构
组2:用bootstrapping
组3:+其他信息
- 组1:仅用结构
7.2EN-FR
- EN-FR: A Benchmarking Study of Embedding-based Entity Alignment for Knowledge Graphsh
7.3 SRPRS
- SRPRS
- 组1:仅用结构
组2:用bootstrapping
组3:+其他信息
- 组1:仅用结构
7.4 DWY100k
参考文献
部分参考未列出
- A Benchmarking Study of Embedding-based Entity Alignment for Knowledge Graphsh