Musical Heritage Historical Entity Linking

2025-02-13

AI Summary

Entity linking in music historical texts is severely hindered by the widespread absence or sparsity of named entities—especially historical figures, works, and institutions—in mainstream knowledge bases (KBs). Method: We propose an unsupervised, knowledge graph (KG)-enhanced framework that uniquely integrates KG logical constraints with heuristic NIL (Not-In-KB) entity identification, leveraging KG embedding, unsupervised semantic matching, and historical context modeling. Contribution/Results: To support evaluation, we introduce MHERCL—the first fine-grained, human-annotated benchmark for music cultural heritage, covering three underrepresented entity types. Experiments demonstrate that our model consistently outperforms state-of-the-art methods on MHERCL, achieving substantial gains in linking robustness and cross-domain generalization. This validates the efficacy of KG-guided unsupervised learning for historical text entity linking.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Natural Language Processing: Information Extraction
• Machine Learning: Graph-based Machine Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Linking named entities occurring in text to their corresponding entity in a Knowledge Base (KB) is challenging, especially when dealing with historical texts. In this work, we introduce Musical Heritage named Entities Recognition, Classification and Linking (MHERCL), a novel benchmark consisting of manually annotated sentences extrapolated from historical periodicals of the music domain. MHERCL contains named entities under-represented or absent in the most famous KBs. We experiment with several State-of-the-Art models on the Entity Linking (EL) task and show that MHERCL is a challenging dataset for all of them. We propose a novel unsupervised EL model and a method to extend supervised entity linkers by using Knowledge Graphs (KGs) to tackle the main difficulties posed by historical documents. Our experiments reveal that relying on unsupervised techniques and improving models with logical constraints based on KGs and heuristics to predict NIL entities (entities not represented in the KB of reference) results in better EL performance on historical documents.

Problem

• Linking historical musical named entities
• Challenges in historical text entity linking
• Improving entity linking with unsupervised techniques

Innovation

• Unsupervised Entity Linking model
• Knowledge Graphs enhance linkers
• Heuristics predict NIL entities