Regresar

Ranking knowledge graphs by capturing knowledge about languages and labels

Abstract:

Capturing knowledge about the mulitilinguality of a knowledge graph is of supreme importance to understand its applicability across multiple languages. Several metrics have been proposed for describing mulitilinguality at the level of a whole knowledge graph. Albeit enabling the understanding of the ecosystem of knowledge graphs in terms of the utilized languages, they are unable to capture a fine-grained description of the languages in which the different entities and properties of the knowledge graph are represented. This lack of representation prevents the comparison of existing knowledge graphs in order to decide which are the most appropriate for a multilingual application. In this work, we approach the problem of ranking knowledge graphs based on their language features and propose LINGVO, a framework able to capture mulitilinguality at different levels of granularity. Grounded in knowledge graph descriptions, LINGVO is, additionally, able to solve the problem of ranking knowledge graphs according to a degree of mulitilinguality of the represented entities. We have empirically studied the effectiveness of LINGVO in a benchmark of queries to be executed against existing knowledge graphs. The observed results provide evidence that LINGVO captures the mulitilinguality of the studied knowledge graphs similarly than a crowd-sourced gold standard.