A Simplified Benchmark for Non-ambiguous Explanations of Knowledge Graph Link Prediction using Relational Graph Convolutional Networks - Archive ouverte HAL Access content directly
Poster Communications Year : 2021

A Simplified Benchmark for Non-ambiguous Explanations of Knowledge Graph Link Prediction using Relational Graph Convolutional Networks

Abstract

Relational Graph Convolutional Networks (RGCNs) identify relationships within a Knowledge Graph to learn real-valued embeddings for each node and edge. Recently, researchers have proposed explanation methods to interpret the predictions of these black-box models. However, comparisons across explanation methods is difficult without a common dataset and standard evaluation metrics to evaluate the explanations. In this paper, we propose a method, including two datasets (Royalty-20k and Royalty-30k), to benchmark explanation methods on the task of explainable link prediction using Graph Neural Networks. We report the results of state-of-the-art explanation methods for RGCNs.
Fichier principal
Vignette du fichier
paper326.pdf (181.63 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03339562 , version 1 (09-09-2021)
hal-03339562 , version 2 (07-10-2021)
hal-03339562 , version 3 (18-11-2021)

Identifiers

  • HAL Id : hal-03339562 , version 3

Cite

Nicholas Halliwell, Fabien Gandon, Freddy Lecue. A Simplified Benchmark for Non-ambiguous Explanations of Knowledge Graph Link Prediction using Relational Graph Convolutional Networks. ISWC 2021 - International Semantic Web Conference, Oct 2021, Troy, United States. Proceedings of International Semantic Web Conference. ⟨hal-03339562v3⟩
288 View
104 Download

Share

Gmail Facebook X LinkedIn More