A Three-stage Neuro-symbolic Recommendation Pipeline for Cultural Heritage Knowledge Graphs

TL;DR

This paper introduces a three-stage neuro-symbolic recommendation pipeline that combines knowledge graph embeddings, approximate nearest-neighbour search, and semantic filtering to improve recommendations in cultural heritage knowledge graphs, demonstrating effectiveness on a large dataset.

Contribution

The work presents a novel hybrid recommendation pipeline integrating multiple techniques for cultural heritage data, with comprehensive evaluation and expert validation.

Findings

Effective recommendations despite sparse data

Explainability of the recommendations

Successful evaluation on a large cultural heritage dataset

Abstract

The growing volume of digital cultural heritage resources highlights the need for advanced recommendation methods capable of interpreting semantic relationships between heterogeneous data entities. This paper presents a complete methodology for implementing a hybrid recommendation pipeline integrating knowledge-graph embeddings, approximate nearest-neighbour search, and SPARQL-driven semantic filtering. The work is evaluated on the JUHMP (Jagiellonian University Heritage Metadata Portal) knowledge graph developed within the CHExRISH project, which at the time of experimentation contained ${\approx}3.2$M RDF triples describing people, events, objects, and historical relations affiliated with the Jagiellonian University (Krak\'{o}w, PL). We evaluate four embedding families (TransE, ComplEx, ConvE, CompGCN) and perform hyperparameter selection for ComplEx and HNSW. Then, we present and evaluate the final three-stage neuro-symbolic recommender. Despite sparse and heterogeneous metadata, the approach produces useful and explainable recommendations, which were also proven with expert evaluation.