# Knowledge Graph Fact Prediction via Knowledge-Enriched Tensor Factorization

## TL;DR

This paper introduces novel tensor-based methods for embedding knowledge graphs that incorporate prior knowledge and demonstrate significant improvements in fact prediction accuracy across multiple datasets.

## Contribution

The paper proposes a family of tensor factorization methods for knowledge graph embedding that effectively utilize background knowledge and provide a convergent algorithm.

## Key findings

- Achieved 5-50% improvement over state-of-the-art methods.
- Tensor models perform well with high average degree entities.
- Constraint models excel with few similar relations, regularization models with diverse relations.

## Abstract

We present a family of novel methods for embedding knowledge graphs into real-valued tensors. These tensor-based embeddings capture the ordered relations that are typical in the knowledge graphs represented by semantic web languages like RDF. Unlike many previous models, our methods can easily use prior background knowledge provided by users or extracted automatically from existing knowledge graphs. In addition to providing more robust methods for knowledge graph embedding, we provide a provably-convergent, linear tensor factorization algorithm. We demonstrate the efficacy of our models for the task of predicting new facts across eight different knowledge graphs, achieving between 5% and 50% relative improvement over existing state-of-the-art knowledge graph embedding techniques. Our empirical evaluation shows that all of the tensor decomposition models perform well when the average degree of an entity in a graph is high, with constraint-based models doing better on graphs with a small number of highly similar relations and regularization-based models dominating for graphs with relations of varying degrees of similarity.