Explainable Link Prediction for Emerging Entities in Knowledge Graphs

• URL: http://arxiv.org/abs/2005.00637v2
• Date: Fri, 25 Sep 2020 13:38:29 GMT
• Title: Explainable Link Prediction for Emerging Entities in Knowledge Graphs
• Authors: Rajarshi Bhowmik and Gerard de Melo
• Abstract summary: Cross-domain knowledge graphs suffer from inherent incompleteness and sparsity. Link prediction can alleviate this by inferring a target entity, given a source entity and a query relation. We propose an inductive representation learning framework that is able to learn representations of previously unseen entities.
• Score: 44.87285668747474
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite their large-scale coverage, cross-domain knowledge graphs invariably suffer from inherent incompleteness and sparsity. Link prediction can alleviate this by inferring a target entity, given a source entity and a query relation. Recent embedding-based approaches operate in an uninterpretable latent semantic vector space of entities and relations, while path-based approaches operate in the symbolic space, making the inference process explainable. However, these approaches typically consider static snapshots of the knowledge graphs, severely restricting their applicability for evolving knowledge graphs with newly emerging entities. To overcome this issue, we propose an inductive representation learning framework that is able to learn representations of previously unseen entities. Our method finds reasoning paths between source and target entities, thereby making the link prediction for unseen entities interpretable and providing support evidence for the inferred link.

Related papers

• Inference over Unseen Entities, Relations and Literals on Knowledge Graphs [1.7474352892977463]
  knowledge graph embedding models have been successfully applied in the transductive setting to tackle various challenging tasks. We propose the attentive byte-pair encoding layer (BytE) to construct a triple embedding from a sequence of byte-pair encoded subword units of entities and relations. BytE leads to massive feature reuse via weight tying, since it forces a knowledge graph embedding model to learn embeddings for subword units instead of entities and relations directly.
  arXiv  Detail & Related papers  (2024-10-09T10:20:54Z)
• You Only Transfer What You Share: Intersection-Induced Graph Transfer Learning for Link Prediction [79.15394378571132]
  We investigate a previously overlooked phenomenon: in many cases, a densely connected, complementary graph can be found for the original graph. The denser graph may share nodes with the original graph, which offers a natural bridge for transferring selective, meaningful knowledge. We identify this setting as Graph Intersection-induced Transfer Learning (GITL), which is motivated by practical applications in e-commerce or academic co-authorship predictions.
  arXiv  Detail & Related papers  (2023-02-27T22:56:06Z)
• Learning Attention-based Representations from Multiple Patterns for Relation Prediction in Knowledge Graphs [2.4028383570062606]
  AEMP is a novel model for learning contextualized representations by acquiring entities' context information. AEMP either outperforms or competes with state-of-the-art relation prediction methods.
  arXiv  Detail & Related papers  (2022-06-07T10:53:35Z)
• Leveraging Unlabeled Data for Entity-Relation Extraction through Probabilistic Constraint Satisfaction [54.06292969184476]
  We study the problem of entity-relation extraction in the presence of symbolic domain knowledge. Our approach employs semantic loss which captures the precise meaning of a logical sentence. With a focus on low-data regimes, we show that semantic loss outperforms the baselines by a wide margin.
  arXiv  Detail & Related papers  (2021-03-20T00:16:29Z)
• Knowledge Graph Embedding using Graph Convolutional Networks with Relation-Aware Attention [3.803929794912623]
  Knowledge graph embedding methods learn embeddings of entities and relations in a low dimensional space. Various graph convolutional network methods have been proposed which use different types of information to learn the features of entities and relations. We propose a relation-aware graph attention model that leverages relation information to compute different weights to the neighboring nodes for learning embeddings of entities and relations.
  arXiv  Detail & Related papers  (2021-02-14T17:19:44Z)
• Joint Semantics and Data-Driven Path Representation for Knowledge Graph Inference [60.048447849653876]
  We propose a novel joint semantics and data-driven path representation that balances explainability and generalization in the framework of KG embedding. Our proposed model is evaluated on two classes of tasks: link prediction and path query answering task.
  arXiv  Detail & Related papers  (2020-10-06T10:24:45Z)
• Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Prediction [69.1473775184952]
  We introduce a realistic problem of few-shot out-of-graph link prediction. We tackle this problem with a novel transductive meta-learning framework. We validate our model on multiple benchmark datasets for knowledge graph completion and drug-drug interaction prediction.
  arXiv  Detail & Related papers  (2020-06-11T17:42:46Z)
• Relational Message Passing for Knowledge Graph Completion [78.47976646383222]
  We propose a relational message passing method for knowledge graph completion. It passes relational messages among edges iteratively to aggregate neighborhood information. Results show our method outperforms stateof-the-art knowledge completion methods by a large margin.
  arXiv  Detail & Related papers  (2020-02-17T03:33:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.