Publication Type

PhD Dissertation

Version

publishedVersion

Publication Date

11-2025

Abstract

Knowledge graphs (KGs) are powerful tools for structuring factual knowledge into relational triples, yet their practical utility is often adversely affected by data sparsity. Many entities and relations are associated with only a few observations, which limits the quality of learned embeddings and weakens generalization in downstream tasks. The problem of sparsity led to two interrelated challenges. Firstly, it restricts the informativeness of training samples: positive examples are scarce, and conventional negative sampling often produces trivial or redundant negatives that resulting in limited guidance. Secondly, in few-shot relation learning scenarios, sparsity worsens distribution shifts between training and test relations, as models trained on base relations find it hard to adapt to novel, sparsely observed relations. To address these challenges, this dissertation develops solutions from two complementary perspectives: on one hand, data-level methods that enrich supervision, relation coverage, and on the other hand, model-level methods that improve adaptations to downstream relations under distribution shifts.

Keywords

Knowledge Graph Embedding, Multimodal datasets, Meta-learning, Adversarial Learning

Degree Awarded

PhD in Computer Science

Discipline

Databases and Information Systems | Graphics and Human Computer Interfaces

Supervisor(s)

FANG, Yuan

First Page

1

Last Page

112

Publisher

Singapore Management University

City or Country

Singapore

Copyright Owner and License

Author

Download

Share

COinS