SAFRAN: An interpretable, rule-based link prediction method outperforming embedding models

09/16/2021
by   Simon Ott, et al.
0

Neural embedding-based machine learning models have shown promise for predicting novel links in knowledge graphs. Unfortunately, their practical utility is diminished by their lack of interpretability. Recently, the fully interpretable, rule-based algorithm AnyBURL yielded highly competitive results on many general-purpose link prediction benchmarks. However, current approaches for aggregating predictions made by multiple rules are affected by redundancies. We improve upon AnyBURL by introducing the SAFRAN rule application framework, which uses a novel aggregation approach called Non-redundant Noisy-OR that detects and clusters redundant rules prior to aggregation. SAFRAN yields new state-of-the-art results for fully interpretable link prediction on the established general-purpose benchmarks FB15K-237, WN18RR and YAGO3-10. Furthermore, it exceeds the results of multiple established embedding-based algorithms on FB15K-237 and WN18RR and narrows the gap between rule-based and embedding-based algorithms on YAGO3-10.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/10/2020

Scalable and interpretable rule-based link prediction for large heterogeneous knowledge graphs

Neural embedding-based machine learning models have shown promise for pr...
research
02/14/2023

Effects of Locality and Rule Language on Explanations for Knowledge Graph Embeddings

Knowledge graphs (KGs) are key tools in many AI-related tasks such as re...
research
04/03/2020

A rigorous method to compare interpretability of rule-based algorithms

Interpretability is becoming increasingly important in predictive model ...
research
09/09/2019

Combination of Unified Embedding Model and Observed Features for Knowledge Graph Completion

Knowledge graphs are useful for many artificial intelligence tasks but o...
research
09/09/2019

Combination of Embedding Models and Collaboration with Observed Feature Models for Knowledge Graph Completion

Knowledge graphs are useful for many artificial intelligence tasks but o...
research
07/01/2020

TransINT: Embedding Implication Rules in Knowledge Graphs with Isomorphic Intersections of Linear Subspaces

Knowledge Graphs (KG), composed of entities and relations, provide a str...
research
02/10/2018

Forget Markov Logic: Efficient Mining of Prediction Rules in Large Graphs

Graph representations of large knowledge bases may comprise billions of ...

Please sign up or login with your details

Forgot password? Click here to reset