Knowledge Graph Fact Prediction via Knowledge-Enriched Tensor Factorization
This work addresses the challenge of knowledge graph completion for applications in semantic web and AI, representing an incremental advancement with specific performance gains.
The authors tackled the problem of predicting new facts in knowledge graphs by introducing a family of tensor factorization methods that incorporate prior background knowledge, achieving relative improvements of 5% to 50% over state-of-the-art techniques across eight knowledge graphs.
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.