Learning Triple Embeddings from Knowledge Graphs

Fionda, Valeria; Pirrò, Giuseppe

doi:10.1609/aaai.v34i04.5800

Cited by 15 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In information retrieval, a neural fact contextualization method has been proposed to rank a set of candidate facts for a given triplet [33]. Also, a way of representing a triplet in an embedding space is studied by considering the concept of a line graph [9]. Recently, ATOMIC [30] has been proposed to provide commonsense knowledge for if-then reasoning, whereas ASER [39] has been proposed to construct an eventuality knowledge graph.…”

Section: Related Workmentioning

confidence: 99%

Learning Representations of Bi-level Knowledge Graphs for Reasoning beyond Link Prediction

Chung¹,

Whang²

2023

Preprint

View full text Add to dashboard Cite

Knowledge graphs represent known facts using triplets. While existing knowledge graph embedding methods only consider the connections between entities, we propose considering the relationships between triplets. For example, let us consider two triplets T1 and T2 where T1 is (Academy Awards, Nominates, Avatar) and T2 is (Avatar, Wins, Academy Awards). Given these two base-level triplets, we see that T1 is a prerequisite for T2. In this paper, we define a higher-level triplet to represent a relationship between triplets, e.g., T1, PrerequisiteFor, T2 where Prerequisite-For is a higher-level relation. We define a bi-level knowledge graph that consists of the base-level and the higher-level triplets. We also propose a data augmentation strategy based on the random walks on the bi-level knowledge graph to augment plausible triplets. Our model called BiVE learns embeddings by taking into account the structures of the base-level and the higher-level triplets, with additional consideration of the augmented triplets. We propose two new tasks: triplet prediction and conditional link prediction. Given a triplet T1 and a higher-level relation, the triplet prediction predicts a triplet that is likely to be connected to T1 by the higher-level relation, e.g., T1, PrerequisiteFor, ? . The conditional link prediction predicts a missing entity in a triplet conditioned on another triplet, e.g., T1, PrerequisiteFor, (Avatar, Wins, ?) . Experimental results show that BiVE significantly outperforms all other methods in the two new tasks and the typical baselevel link prediction in real-world bi-level knowledge graphs.

show abstract

Section: Related Workmentioning

confidence: 99%

Learning Representations of Bi-level Knowledge Graphs for Reasoning beyond Link Prediction

Chung¹,

Whang²

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…This model generates embeddings in which entities that are in lower-order neighbourhoods of one another are more proximal in the embedding space. Being one of the first and most popular embedding approaches, it has received significant research attention and extensions, such as KG2Vec [22], Triple2Vec [23], and RDF2Vec_oa [24]. GCNs have been extended to knowledge graphs with the Relational Graph Convolution Network (R-GCN) [3], which aggregates predicate-specific convolutions of the original model.…”

Section: Related Workmentioning

confidence: 99%

Probabilistic Coarsening for Knowledge Graph Embeddings

Pietrasik

Reformat

2023

Axioms

View full text Add to dashboard Cite

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.

show abstract

“…In information retrieval, a neural fact contextualization method has been proposed to rank a set of candidate facts for a given triplet (Voskarides et al 2018). Also, a way of representing a triplet in an embedding space is studied by considering the concept of a line graph (Fionda and Pirrò 2020). Recently, ATOMIC (Sap et al 2019) has been proposed to provide commonsense knowledge for if-then reasoning, whereas ASER (Zhang et al 2020) has been proposed to construct an eventuality knowledge graph.…”

Section: Related Workmentioning

confidence: 99%

Learning Representations of Bi-level Knowledge Graphs for Reasoning beyond Link Prediction

Chung

Whang

2023

AAAI

View full text Add to dashboard Cite

Knowledge graphs represent known facts using triplets. While existing knowledge graph embedding methods only consider the connections between entities, we propose considering the relationships between triplets. For example, let us consider two triplets T1 and T2 where T1 is (Academy_Awards, Nominates, Avatar) and T2 is (Avatar, Wins, Academy_Awards). Given these two base-level triplets, we see that T1 is a prerequisite for T2. In this paper, we define a higher-level triplet to represent a relationship between triplets, e.g., where PrerequisiteFor is a higher-level relation. We define a bi-level knowledge graph that consists of the base-level and the higher-level triplets. We also propose a data augmentation strategy based on the random walks on the bi-level knowledge graph to augment plausible triplets. Our model called BiVE learns embeddings by taking into account the structures of the base-level and the higher-level triplets, with additional consideration of the augmented triplets. We propose two new tasks: triplet prediction and conditional link prediction. Given a triplet T1 and a higher-level relation, the triplet prediction predicts a triplet that is likely to be connected to T1 by the higher-level relation, e.g., . The conditional link prediction predicts a missing entity in a triplet conditioned on another triplet, e.g., . Experimental results show that BiVE significantly outperforms all other methods in the two new tasks and the typical base-level link prediction in real-world bi-level knowledge graphs.

show abstract

Learning Triple Embeddings from Knowledge Graphs

Cited by 15 publications

References 8 publications

Learning Representations of Bi-level Knowledge Graphs for Reasoning beyond Link Prediction

Learning Representations of Bi-level Knowledge Graphs for Reasoning beyond Link Prediction

Probabilistic Coarsening for Knowledge Graph Embeddings

Learning Representations of Bi-level Knowledge Graphs for Reasoning beyond Link Prediction

Contact Info

Product

Resources

About