Using Graph-Pattern Association Rules On Yago Knowledge Base

TL;DR

This paper introduces Graph-Pattern Association Rules (GPARs) for knowledge bases, demonstrating their effectiveness on Yago by generating over a thousand rules with high confidence and efficient computation.

Contribution

It extends association rule techniques to graph patterns in knowledge bases and proposes algorithms for rule extraction and creation, improving confidence and computation time.

Findings

Generated 1114 association rules from Yago.

Standard confidence achieved 50.18%, outperforming PCA confidence.

Faster computation time for standard confidence than PCA confidence.

Abstract

We propose the use of Graph-Pattern Association Rules (GPARs) on the Yago knowledge base. Extending association rules for itemsets, GPARS can help to discover regularities between entities in knowledge bases. A rule-generated graph pattern (RGGP) algorithm was used for extracting rules from the Yago knowledge base and a graph-pattern association rules algorithm for creating association rules. Our research resulted in 1114 association rules, where the value of standard confidence at 50.18% was better than partial completeness assumption (PCA) confidence at 49.82%. Besides that the computation time for standard confidence was also better than for PCA confidence