2019
DOI: 10.1609/aaai.v33i01.33013363
|View full text |Cite
|
Sign up to set email alerts
|

Embedding Uncertain Knowledge Graphs

Abstract: Embedding models for deterministic Knowledge Graphs (KG) have been extensively studied, with the purpose of capturing latent semantic relations between entities and incorporating the structured knowledge they contain into machine learning. However, there are many KGs that model uncertain knowledge, which typically model the inherent uncertainty of relations facts with a confidence score, and embedding such uncertain knowledge represents an unresolved challenge. The capturing of uncertain knowledge will benefit… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
73
0

Year Published

2020
2020
2022
2022

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 94 publications
(83 citation statements)
references
References 24 publications
0
73
0
Order By: Relevance
“…To enhance the learning ability of the evaluator, we use unknown facts to precipitate the training process. We introduce heuristic rules following the practice of Chen et al [11] and use the probabilistic soft logic to obtain unknown facts. However, we find that the rules mined from UKG are inherently uncertain, inferring inaccurate unknown facts unavoidably.…”
Section: ) Probabilistic Soft Logic Enhancement Methodsmentioning
confidence: 99%
See 4 more Smart Citations
“…To enhance the learning ability of the evaluator, we use unknown facts to precipitate the training process. We introduce heuristic rules following the practice of Chen et al [11] and use the probabilistic soft logic to obtain unknown facts. However, we find that the rules mined from UKG are inherently uncertain, inferring inaccurate unknown facts unavoidably.…”
Section: ) Probabilistic Soft Logic Enhancement Methodsmentioning
confidence: 99%
“…Therefore, the application of URGE to UKG has certain limitations. Another recent model proposed by Chen et al in 2019 is UKGE [11]. UKGE defines the plausibility score for each triple, and proposes two variants.In addition, UKGE defines the calculation method of basic logic operations (logical conjunction ∧, disjunction ∨, and negation ¬) based on the firstorder logic rules, and correctly estimates the unknown triples and corresponding confidence in the uncertain knowledge base.…”
Section: B Uncertain Knowledge Graphmentioning
confidence: 99%
See 3 more Smart Citations