MELArt: Multimodal Entity Linking Evaluation Dataset for Art (Version 1.0)

This dataset encapsulates various entities unique to art, presenting a robust tool tailored to the creation of multi-modal art datasets. We also apply distinct metrics that assess ambiguity and diversity in the data, offering a thorough evaluation of our dataset. We introduce an automated process that facilitates the generation of art datasets, harnessing data from multiple sources (ArtPedia, Wikidata and Wikimedia Commons) to ensure reliability and comprehensiveness. Furthermore, our paper provides extensive statistical data, delineates best practices for the integration of art datasets, and presents a detailed performance analysis of entity linking systems, when applied to domain-specific datasets.

本数据集涵盖了艺术领域独有的各类实体，提供了一款专为多模态艺术数据集构建打造的健壮工具。本研究同时采用了用于评估数据歧义性与多样性的专属评估指标，对本数据集开展了全面评测。本研究提出了一种自动化流程以助力艺术数据集的生成，该流程依托ArtPedia、维基数据（Wikidata）与维基共享资源（Wikimedia Commons）等多源数据，确保了数据集的可靠性与全面性。此外，本文还提供了详尽的统计数据，阐明了艺术数据集集成的最佳实践方案，并针对应用于领域特定数据集的实体链接（entity linking）系统开展了细致的性能分析。