FLORA: Unsupervised Knowledge Graph Alignment by Fuzzy Logic

TL;DR

FLORA tackles the unsupervised holistic alignment of entities and relations across knowledge graphs using a principled fuzzy logic framework. It formulates a recursive fuzzy inference system with fixed-point convergence, leveraging relation lists and functionality signals to align both instances and relations, including subrelations, while handling dangling entities. The approach achieves state-of-the-art performance on major EA and holistic KG benchmarks, with strong interpretability: the most influential rule explains each alignment. FLORA remains effective without training data and scales to large datasets, offering practical utility for knowledge fusion and cross-graph reasoning.

Abstract

Knowledge graph alignment is the task of matching equivalent entities (that is, instances and classes) and relations across two knowledge graphs. Most existing methods focus on pure entity-level alignment, computing the similarity of entities in some embedding space. They lack interpretable reasoning and need training data to work. In this paper, we propose FLORA, a simple yet effective method that (1) is unsupervised, i.e., does not require training data, (2) provides a holistic alignment for entities and relations iteratively, (3) is based on fuzzy logic and thus delivers interpretable results, (4) provably converges, (5) allows dangling entities, i.e., entities without a counterpart in the other KG, and (6) achieves state-of-the-art results on major benchmarks.

Figures (3)

• Figure 1: An excerpt of DBpedia and Wikidata KGs. Note the asymmetric mapping of the relation parent to the sub-relation father.
• Figure 2: Knowledge Graph Alignment with FLORA.
• Figure 3: Sensitivity to the benefit of doubt $\alpha$

Theorems & Definitions (4)

• definition thmcounterdefinition: Simple Positive FIS
• definition thmcounterdefinition: Recursive FIS
• theorem thmcountertheorem
• proof