On Training Knowledge Graph Embedding Models
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F21%3A00121336" target="_blank" >RIV/00216224:14330/21:00121336 - isvavai.cz</a>
Result on the web
<a href="https://www.mdpi.com/2078-2489/12/4/147" target="_blank" >https://www.mdpi.com/2078-2489/12/4/147</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.3390/info12040147" target="_blank" >10.3390/info12040147</a>
Alternative languages
Result language
angličtina
Original language name
On Training Knowledge Graph Embedding Models
Original language description
Knowledge graph embedding (KGE) models have become popular means for making discoveries in knowledge graphs (e.g., RDF graphs) in an efficient and scalable manner. The key to success of these models is their ability to learn low-rank vector representations for knowledge graph entities and relations. Despite the rapid development of KGE models, state-of-the-art approaches have mostly focused on new ways to represent embeddings interaction functions (i.e., scoring functions). In this paper, we argue that the choice of other training components such as the loss function, hyperparameters and negative sampling strategies can also have substantial impact on the model efficiency. This area has been rather neglected by previous works so far and our contribution is towards closing this gap by a thorough analysis of possible choices of training loss functions, hyperparameters and negative sampling techniques. We finally investigate the effects of specific choices on the scalability and accuracy of knowledge graph embedding models.
Czech name
—
Czech description
—
Classification
Type
J<sub>imp</sub> - Article in a specialist periodical, which is included in the Web of Science database
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
O - Projekt operacniho programu
Others
Publication year
2021
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Name of the periodical
Information
ISSN
2078-2489
e-ISSN
—
Volume of the periodical
12
Issue of the periodical within the volume
4
Country of publishing house
CH - SWITZERLAND
Number of pages
19
Pages from-to
147
UT code for WoS article
000643062300001
EID of the result in the Scopus database
2-s2.0-85104536343