Species, emotion and genre in Japanese animation: a linked open data analysis of anime animal characters

Purpose This study aims to examine whether linked open data can support large-scale analysis of animal-character species, genre, and emotional archetype in Japanese animation, while clarifying the relevance of anime metadata for digital library research. Design/methodology/approach Character-level records were assembled through the Japanese Visual Media Graph (JVMG), which integrates Anime Characters Database, Visual Novel Database and related sources. Species labels were harmonized, mapped to anime genre facets and, where available, compared with Dere-type emotional annotations. A Web of Science/VOSviewer map of anime-genre scholarship contextualized the study. Findings Animal species are not randomly distributed across genres. Cat characters cluster in relational and mood-driven genres, while dog characters appear more strongly in action-oriented and ethically charged genres. Secondary species concentrate in fantasy, supernatural and comedic settings and show thinner emotional annotation. Practical implications The paper demonstrates how linked open data aggregation enables reusable, scalable character analysis applicable to other audiovisual collections. Originality/value The study connects cultural analytics, anime studies and digital-library metadata by showing how linked open data can transform dispersed popular culture records into reusable evidence for genre discovery and character typology.