English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Using knowledge graph embedding models (KGEMs) is a popular approach forpredicting links in knowledge graphs (KGs). Traditionally, the performance ofKGEMs for link prediction is assessed using rank-based metrics, which evaluatetheir ability to give high scores to ground-truth entities. However, theliterature claims that the KGEM evaluation procedure would benefit from addingsupplementary dimensions to assess. That is why, in this paper, we extend ourpreviously introduced metric Sem@K that measures the capability of models topredict valid entities w.r.t. domain and range constraints. In particular, weconsider a broad range of KGs and take their respective characteristics intoaccount to propose different versions of Sem@K. We also perform an extensivestudy to qualify the abilities of KGEMs as measured by our metric. Ourexperiments show that Sem@K provides a new perspective on KGEM quality. Itsjoint analysis with rank-based metrics offers different conclusions on thepredictive power of models. Regarding Sem@K, some KGEMs are inherently betterthan others, but this semantic superiority is not indicative of theirperformance w.r.t. rank-based metrics. In this work, we generalize conclusionsabout the relative performance of KGEMs w.r.t. rank-based and semantic-orientedmetrics at the level of families of models. The joint analysis of theaforementioned metrics gives more insight into the peculiarities of each model.This work paves the way for a more comprehensive evaluation of KGEM adequacyfor specific downstream tasks.

Sem@$K$: Is my knowledge graph embedding model semantic-aware?