Deep Sparse Latent Feature Models for Knowledge Graph Completion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Haotian, Zhang, Rui, Wang, Lingzhi, Yu, Bin, Wang, Youwei, Wei, Yuliang, Wang, Kai, Da Xu, Richard Yi, Wang, Bailing
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918057218146304
author Li, Haotian
Zhang, Rui
Wang, Lingzhi
Yu, Bin
Wang, Youwei
Wei, Yuliang
Wang, Kai
Da Xu, Richard Yi
Wang, Bailing
author_facet Li, Haotian
Zhang, Rui
Wang, Lingzhi
Yu, Bin
Wang, Youwei
Wei, Yuliang
Wang, Kai
Da Xu, Richard Yi
Wang, Bailing
contents Recent advances in knowledge graph completion (KGC) have emphasized text-based approaches to navigate the inherent complexities of large-scale knowledge graphs (KGs). While these methods have achieved notable progress, they frequently struggle to fully incorporate the global structural properties of the graph. Stochastic blockmodels (SBMs), especially the latent feature relational model (LFRM), offer robust probabilistic frameworks for identifying latent community structures and improving link prediction. This paper presents a novel probabilistic KGC framework utilizing sparse latent feature models, optimized via a deep variational autoencoder (VAE). Our proposed method dynamically integrates global clustering information with local textual features to effectively complete missing triples, while also providing enhanced interpretability of the underlying latent structures. Extensive experiments on four benchmark datasets with varying scales demonstrate the significant performance gains achieved by our method.
format Preprint
id arxiv_https___arxiv_org_abs_2411_15694
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Deep Sparse Latent Feature Models for Knowledge Graph Completion
Li, Haotian
Zhang, Rui
Wang, Lingzhi
Yu, Bin
Wang, Youwei
Wei, Yuliang
Wang, Kai
Da Xu, Richard Yi
Wang, Bailing
Computation and Language
Recent advances in knowledge graph completion (KGC) have emphasized text-based approaches to navigate the inherent complexities of large-scale knowledge graphs (KGs). While these methods have achieved notable progress, they frequently struggle to fully incorporate the global structural properties of the graph. Stochastic blockmodels (SBMs), especially the latent feature relational model (LFRM), offer robust probabilistic frameworks for identifying latent community structures and improving link prediction. This paper presents a novel probabilistic KGC framework utilizing sparse latent feature models, optimized via a deep variational autoencoder (VAE). Our proposed method dynamically integrates global clustering information with local textual features to effectively complete missing triples, while also providing enhanced interpretability of the underlying latent structures. Extensive experiments on four benchmark datasets with varying scales demonstrate the significant performance gains achieved by our method.
title Deep Sparse Latent Feature Models for Knowledge Graph Completion
topic Computation and Language
url https://arxiv.org/abs/2411.15694