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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Knowledge graph (KG) embedding is widely used in many downstream applicationsusing KGs. Generally, since KGs contain only ground truth triples, it isnecessary to construct arbitrary negative samples for representation learningof KGs. Recently, various methods for sampling high-quality negatives have beenstudied because the quality of negative triples has great effect on KGembedding. In this paper, we propose a novel method called Entity AwareNegative Sampling (EANS), which is able to sample negative entities resemble topositive one by adopting Gaussian distribution to the aligned entity indexspace. Additionally, we introduce auxiliary loss for false negative predictionthat can alleviate the impact of the sampled false negative triples. Theproposed method can generate high-quality negative samples regardless ofnegative sample size and effectively mitigate the influence of false negativesamples. The experimental results on standard benchmarks show that our EANSoutperforms existing the state-of-the-art methods of negative sampling onseveral knowledge graph embedding models. Moreover, the proposed methodachieves competitive performance even when the number of negative samples islimited to only one.

Entity Aware Negative Sampling with Auxiliary Loss of False Negative Prediction for Knowledge Graph Embedding