Evaluating the Impact of Knowledge Graph Context on Entity Disambiguation Models

TL;DR

This paper demonstrates that incorporating knowledge graph context, specifically from Wikidata, into pretrained transformer models significantly enhances their performance on named entity disambiguation tasks for Wikipedia, showing the generalizability of the approach.

Contribution

The paper introduces a method to integrate Wikidata-based KG context into transformer models, improving NED performance and demonstrating its applicability to Wikipedia.

Findings

KG context improves NED accuracy over baselines

The approach generalizes well to Wikipedia

Transformer models with KG context outperform vanilla models

Abstract

Pretrained Transformer models have emerged as state-of-the-art approaches that learn contextual information from text to improve the performance of several NLP tasks. These models, albeit powerful, still require specialized knowledge in specific scenarios. In this paper, we argue that context derived from a knowledge graph (in our case: Wikidata) provides enough signals to inform pretrained transformer models and improve their performance for named entity disambiguation (NED) on Wikidata KG. We further hypothesize that our proposed KG context can be standardized for Wikipedia, and we evaluate the impact of KG context on state-of-the-art NED model for the Wikipedia knowledge base. Our empirical results validate that the proposed KG context can be generalized (for Wikipedia), and providing KG context in transformer architectures considerably outperforms the existing baselines, including the vanilla transformer models.