RECON: Relation Extraction using Knowledge Graph Context in a Graph Neural Network
This addresses the challenge of improving relation extraction accuracy for natural language processing applications by leveraging knowledge graphs, representing an incremental advance with specific gains.
The paper tackles the problem of relation extraction from sentences by aligning them to a knowledge graph, using a graph neural network to incorporate KG context, resulting in significant performance improvements over state-of-the-art methods, such as an F1 score of 87.23 vs. 82.29 on Wikidata and 87.5(P@10) vs. 81.3(P@10) on NYT Freebase.
In this paper, we present a novel method named RECON, that automatically identifies relations in a sentence (sentential relation extraction) and aligns to a knowledge graph (KG). RECON uses a graph neural network to learn representations of both the sentence as well as facts stored in a KG, improving the overall extraction quality. These facts, including entity attributes (label, alias, description, instance-of) and factual triples, have not been collectively used in the state of the art methods. We evaluate the effect of various forms of representing the KG context on the performance of RECON. The empirical evaluation on two standard relation extraction datasets shows that RECON significantly outperforms all state of the art methods on NYT Freebase and Wikidata datasets. RECON reports 87.23 F1 score (Vs 82.29 baseline) on Wikidata dataset whereas on NYT Freebase, reported values are 87.5(P@10) and 74.1(P@30) compared to the previous baseline scores of 81.3(P@10) and 63.1(P@30).