Semantic Alignment of Multilingual Knowledge Graphs via Contextualized Vector Projections

TL;DR

This paper introduces a multilingual ontology alignment system that leverages contextualized embeddings from a transformer model and cosine similarity to effectively identify cross-lingual entity matches, significantly improving alignment accuracy.

Contribution

It proposes a novel approach combining contextually enriched descriptions and transformer-based embeddings for cross-lingual ontology alignment, achieving a 16% improvement over the baseline.

Findings

Achieved 71% F1 score on OAEI-2022 multifarm track

Improved recall to 78% and precision to 65%

Demonstrated effectiveness of contextualized embeddings in capturing cross-lingual similarities

Abstract

The paper presents our work on cross-lingual ontology alignment system which uses embedding based cosine similarity matching. The ontology entities are made contextually richer by creating descriptions using novel techniques. We use a fine-tuned transformer based multilingual model for generating better embeddings. We use cosine similarity to find positive ontology entities pairs and then apply threshold filtering to retain only highly similar entities. We have evaluated our work on OAEI-2022 multifarm track. We achieve 71% F1 score (78% recall and 65% precision) on the evaluation dataset, 16% increase from best baseline score. This suggests that our proposed alignment pipeline is able to capture the subtle cross-lingual similarities.