Knowledge Graph Entity Alignment with Graph Convolutional Networks: Lessons Learned

Berrendorf, Max; Faerman, Evgeniy; Melnychuk, Valentyn; Tresp, Volker; Seidl, Thomas

doi:10.1007/978-3-030-45442-5_1

Cited by 13 publications

(15 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…rarely report validation splits. 2 Also, in the published code of the investigated approaches, we could not find any trace of train-validation splits, raising questions about reproducibility and fairness of their comparisons. We thus create a shared split with a test, train, and validation part and extensively tune the model's hyperparameters for each of the dataset/initialization combinations to ensure that they are sufficiently optimized.…”

Section: Introductionmentioning

confidence: 92%

“…It has three subsets, all of which base upon DPedia, and comprise a pair of graphs from different languages. As noted by [2], there exist multiple variations of the dataset, sharing the same entity alignment, but differing in the number of exclusive entities in each graph. The alignments in the datasets are always 1:1 alignments, and due to the construction method for the datasets, exclusive entities do not have relations between them, but only to shared entities.…”

Section: Datasetsmentioning

confidence: 99%

“…As in [2], we extract additional entity alignments from the triple alignments. The graphs contain additional exclusive entities, and there are m:n matchings.…”

Section: Datasetsmentioning

confidence: 99%

“…Moreover, to satisfy diverse information needs, a combination of multiple data sources is often inevitable. Entity Alignment (EA) [2] is the discipline of aligning entities from different KGs. Once aligned, these entities facilitate information transfer between both knowledge bases, or even fusing multiple KGs to a single knowledge base.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Critical Assessment of State-of-the-Art in Entity Alignment

Berrendorf,

Wacker,

Faerman

2020

Preprint

Self Cite

View full text Add to dashboard Cite

In this work, we perform an extensive investigation of two state-of-the-art (SotA) methods for the task of Entity Alignment in Knowledge Graphs. Therefore, we first carefully examine the benchmarking process and identify several shortcomings, which make the results reported in the original works not always comparable. Furthermore, we suspect that it is a common practice in the community to make the hyperparameter optimization directly on a test set, reducing the informative value of reported performance. Thus, we select a representative sample of benchmarking datasets and describe their properties. We also examine different initializations for entity representations since they are a decisive factor for model performance. Furthermore, we use a shared train/validation/test split for a fair evaluation setting in which we evaluate all methods on all datasets. In our evaluation, we make several interesting findings. While we observe that most of the time SotA approaches perform better than baselines, they have difficulties when the dataset contains noise, which is the case in most real-life applications. Moreover, we find out in our ablation study that often different features of SotA method are crucial for good performance than previously assumed. The code is available at https://github.com/mberr/ea-sota-comparison.

show abstract

Section: Introductionmentioning

confidence: 92%

Section: Datasetsmentioning

confidence: 99%

“…As in [2], we extract additional entity alignments from the triple alignments. The graphs contain additional exclusive entities, and there are m:n matchings.…”

Section: Datasetsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Critical Assessment of State-of-the-Art in Entity Alignment

Berrendorf,

Wacker,

Faerman

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Most GNN-based entity alignment methods (Berrendorf et al, 2020;Zhu et al, 2020;Mao et al, 2020) are subject to the following framework: (1) a GNN to learn node representations from graph structure and (2) a margin-based loss to rank the distance between entity pairs. The loss function…”

Section: Gnn-based Entity Alignmentmentioning

confidence: 99%

Cybersecurity Entity Alignment via Masked Graph Attention Networks

Qin¹,

Liao²

2022

Preprint

View full text Add to dashboard Cite

Cybersecurity vulnerability information is often recorded by multiple channels, including government vulnerability repositories, individual-maintained vulnerability-gathering platforms, or vulnerability-disclosure email lists and forums. Integrating vulnerability information from different channels enables comprehensive threat assessment and quick deployment to various security mechanisms. Efforts to automatically gather such information, however, are impeded by the limitations of today's entity alignment techniques. In our study, we annotate the first cybersecuritydomain entity alignment dataset and reveal the unique characteristics of security entities. Based on these observations, we propose the first cybersecurity entity alignment model, CEAM, which equips GNN-based entity alignment with two mechanisms: asymmetric masked aggregation and partitioned attention. Experimental results on cybersecuritydomain entity alignment datasets demonstrate that CEAM significantly outperforms state-ofthe-art entity alignment methods.

show abstract

Active Learning for Entity Alignment

Berrendorf

Faerman

Tresp

2021

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

In this work, we propose a novel framework for the labeling of entity alignments in knowledge graph datasets. Different strategies to select informative instances for the human labeler build the core of our framework . We illustrate how the labeling of entity alignments is different from assigning class labels to single instances and how these differences affect the labeling efficiency. Based on these considerations we propose and evaluate different active and passive learning strategies. One of our main findings is that passive learning approaches, which can be efficiently precomputed and deployed more easily, achieve performance comparable to the active learning strategies. Moreover, we can dynamically learn to combine these scores to obtain an even better heuristic.

show abstract

Knowledge Graph Entity Alignment with Graph Convolutional Networks: Lessons Learned

Cited by 13 publications

References 18 publications

A Critical Assessment of State-of-the-Art in Entity Alignment

A Critical Assessment of State-of-the-Art in Entity Alignment

Cybersecurity Entity Alignment via Masked Graph Attention Networks

Active Learning for Entity Alignment

Contact Info

Product

Resources

About