Building a Knowledge Graph of Chinese Kung Fu Masters From Heterogeneous Bilingual Data

Abstract

Various endeavours into semantic web technologies and ontology engineering have been made within the organisation of cultural data, facilitating public access to digital assets. Although models for conceptualising objects have reached a certain level of maturity, only a few have delved into the tacit roles of individuals in the development of knowledge. Simultaneously, the field of cultural analytics demands practical methods for integrating diverse multilingual materials into a consistent data representation. In this context, our work addresses a human-centred perspective to construct a knowledge graph presenting historical Chinese kung fu masters, aptly named MA2KG (Martial Arts MAsters Knowledge Graph). The data workflow is built upon an established ontology model that describes traditional martial arts. It aggregates information from heterogeneous bilingual sources through direct connections and rule-based inference, incorporating data from English Wikidata and Chinese Baidu Baike and complemented with manual annotations. In addition, we describe our methodology and process in making the dataset available with scripts for reproducing similar agent-mediated contexts, data application and inspection cases are provided to discuss our findings and concerns regarding the use of linked open data strategies.

Keywords

• knowledge graphs
• social networks
• chinese martial arts
• linked open data
• digital humanities