Wikidata has raised signifcant interest among researchers and practitioners, due to some of its prominent features, such as openness, broad coverage, and a large user pool. In spite of the growing body of research about Wikidata, only few studies have focused on the quality of its data and even less on how this is determined by the collaborative patterns within its community. Our work seeks to address this gap. First, we perform an evaluation of some dimensions data quality in Wikidata, namely those related to its external references and to the consistency of its conceptual knowledge. Second, we analyse how editing patterns, specifcally the composition of group of users, determine some aspects of its quality. The results obtained so far highlight a high level of quality of external sources used in Wikidata, in terms of relevance and authoritativeness. Moreover, for what concerns community dynamics, our fndings show that a balanced contribution of human and bot users is benefcial for Wikidata quality.

Started:
2015-04-13
Supervisors:

Publications

Wikidatians Are Born: Paths to Full Participation in a Collaborative Structured Knowledge Base. Alessandro Piscopo, Christopher Phethean, Elena Simperl. HICSS-50. 2017 PDF

What do Wikidata and Wikipedia Have in Common?: An Analysis of their Use of External References. Alessandro Piscopo, Pavlos Vougiouklis, Lucie-Aimee Kaffee, Christopher Phethean, Jonathon Hare, and Elena Simperl. OpenSym 2017: 1:1-1:10 URL PDF

What Makes a Good Collaborative Knowledge Graph: Group Composition and Quality in Wikidata. Alessandro Piscopo, Chris Phethean, and Elena Simperl. Social Informatics 2017: 305-322 URL PDF

Provenance Information in a Collaborative Knowledge Graph: an Evaluation of Wikidata External References. Alessandro Piscopo, Lucie-Aimée Kaffee, Chris Phethean, Elena Simperl. ISWC (1) 2017: 542-558 URL PDF

A Glimpse into Babel: An Analysis of Multilinguality in Wikidata. Lucie-Aimée Kaffee, Alessandro Piscopo, Pavlos Vougiouklis, Elena Simperl, Leslie Carr, Lydia Pintscher. OpenSym 2017: 14:1-14:5 URL PDF