Fast Linear Model for Knowledge Graph Embeddings
arXiv:1710.10881
Abstract
This paper shows that a simple baseline based on a Bag-of-Words (BoW) representation learns surprisingly good knowledge graph embeddings. By casting knowledge base completion and question answering as supervised classification problems, we observe that modeling co-occurences of entities and relations leads to state-of-the-art performance with a training time of a few minutes using the open sourced library fastText.
Submitted AKBC 2017
References in corpus (9)
- Large-scale Simple Question Answering with Memory Networks
- Question Answering with Subgraph Embeddings
- Character-Level Question Answering with Attention
- Modeling Relational Data with Graph Convolutional Networks
- Knowledge Graph Completion via Complex Tensor Factorization
- Simple Question Answering by Attentive Convolutional Neural Network
- CFO: Conditional Focused Neural Question Answering with Large-scale Knowledge Bases
- Complex and Holographic Embeddings of Knowledge Graphs: A Comparison
- Link Prediction using Embedded Knowledge Graphs
Cited by in corpus (9)
- Canonical Tensor Decomposition for Knowledge Base Completion
- Augmenting Neural Machine Translation with Knowledge Graphs
- On Evaluating Embedding Models for Knowledge Base Completion
- RDF2Vec-based Classification of Ontology Alignment Changes
- Highly Efficient Knowledge Graph Embedding Learning with Orthogonal Procrustes Analysis
- Generalizing Tensor Decomposition for N-ary Relational Knowledge Bases
- Embedding Geographic Locations for Modelling the Natural Environment using Flickr Tags and Structured Data
- Injecting Knowledge Base Information into End-to-End Joint Entity and Relation Extraction and Coreference Resolution
- Using Pairwise Occurrence Information to Improve Knowledge Graph Completion on Large-Scale Datasets