MUSEKG: A Knowledge Graph Over Museum Collections

2025-11-19

AI Summary

To address the fragmentation of museum cultural relic data, heterogeneity of metadata, and challenges in unified modeling and querying of multimodal resources, this paper proposes an end-to-end knowledge graph construction framework that synergistically integrates symbolic and neural models. Leveraging typed property graph technology, the framework jointly incorporates natural language processing, structured metadata mapping, and multimodal knowledge graph embedding to achieve unified, interpretable representation and semantic reasoning over heterogeneous metadata, unstructured documents, and multimodal (image–text) artifact data. It supports natural language queries while ensuring semantic consistency and provenance traceability. Experiments on a real-world museum dataset demonstrate that our method significantly outperforms large language model baselines—including zero-shot and few-shot variants—as well as SPARQL-prompting approaches across entity-, relation-, and attribute-level query tasks, validating its effectiveness and practical utility.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Knowledge Representation and Reasoning: Knowledge Representation Languages

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: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Digital transformation in the cultural heritage sector has produced vast yet fragmented collections of artefact data. Existing frameworks for museum information systems struggle to integrate heterogeneous metadata, unstructured documents, and multimodal artefacts into a coherent and queryable form. We present MuseKG, an end-to-end knowledge-graph framework that unifies structured and unstructured museum data through symbolic-neural integration. MuseKG constructs a typed property graph linking objects, people, organisations, and visual or textual labels, and supports natural language queries. Evaluations on real museum collections demonstrate robust performance across queries over attributes, relations, and related entities, surpassing large-language-model zero-shot, few-shot and SPARQL prompt baselines. The results highlight the importance of symbolic grounding for interpretable and scalable cultural heritage reasoning, and pave the way for web-scale integration of digital heritage knowledge.

Problem

• Integrating fragmented museum data into unified knowledge graphs
• Handling heterogeneous metadata and multimodal cultural heritage artefacts
• Enabling natural language queries over museum collections and relationships

Innovation

• Unifies structured and unstructured museum data
• Constructs typed property graph linking entities
• Supports natural language queries over knowledge