Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

NALA: an Effective and Interpretable Entity Alignment Method

Chuanhao Xu, Jingwei Cheng, Fu Zhang

TL;DR

NALA iteratively aligns entities and relations by integrating the conclusions of the inference paths with Non-Axiomatic Logic, and is logically interpretable and extensible by introducing NAL, and thus suitable for various EA settings.

Abstract

Entity alignment (EA) aims to find equivalent entities between two Knowledge Graphs. Existing embedding-based EA methods usually encode entities as embeddings, triples as embeddings' constraint and learn to align the embeddings. However, the details of the underlying logical inference steps among the alignment process are usually omitted, resulting in inadequate inference process. In this paper, we introduce NALA, an entity alignment method that captures three types of logical inference paths with Non-Axiomatic Logic (NAL). Type 1&2 align the entity pairs and type 3 aligns relations. NALA iteratively aligns entities and relations by integrating the conclusions of the inference paths. Our method is logically interpretable and extensible by introducing NAL, and thus suitable for various EA settings. Experimental results show that NALA outperforms state-of-the-art methods in terms of Hits@1, achieving 0.98+ on all three datasets of DBP15K with both supervised and unsupervised settings. We offer a pioneering in-depth analysis of the fundamental principles of entity alignment, approaching the subject from a unified and logical perspective. Our code is available at https://github.com/13998151318/NALA.

NALA: an Effective and Interpretable Entity Alignment Method

TL;DR

NALA iteratively aligns entities and relations by integrating the conclusions of the inference paths with Non-Axiomatic Logic, and is logically interpretable and extensible by introducing NAL, and thus suitable for various EA settings.

Abstract

Entity alignment (EA) aims to find equivalent entities between two Knowledge Graphs. Existing embedding-based EA methods usually encode entities as embeddings, triples as embeddings' constraint and learn to align the embeddings. However, the details of the underlying logical inference steps among the alignment process are usually omitted, resulting in inadequate inference process. In this paper, we introduce NALA, an entity alignment method that captures three types of logical inference paths with Non-Axiomatic Logic (NAL). Type 1&2 align the entity pairs and type 3 aligns relations. NALA iteratively aligns entities and relations by integrating the conclusions of the inference paths. Our method is logically interpretable and extensible by introducing NAL, and thus suitable for various EA settings. Experimental results show that NALA outperforms state-of-the-art methods in terms of Hits@1, achieving 0.98+ on all three datasets of DBP15K with both supervised and unsupervised settings. We offer a pioneering in-depth analysis of the fundamental principles of entity alignment, approaching the subject from a unified and logical perspective. Our code is available at https://github.com/13998151318/NALA.
Paper Structure (41 sections, 3 equations, 8 figures, 6 tables, 3 algorithms)

This paper contains 41 sections, 3 equations, 8 figures, 6 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Knowledge Graph and Entity Alignment
  4. Represent KGs with NAL
  5. Why NAL
  6. Related Work of EA
  7. The Proposed Method
  8. Similarity Inference Module
  9. Type I Path: Align Entities by Triples
  10. Type II Path: Align Entities by Name
  11. Type III Path: Aligning Relations
  12. Matching Module
  13. Unsupervised Learning
  14. Interpretability of NALA
  15. Experiments and Results
  16. ...and 26 more sections

Figures (8)

  • Figure 1: An overview illustration of NALA.
  • Figure 2: NAL formalization of type I path
  • Figure 3: An instance of type I path, fetching from DBP15K zh-en and omitting irrelevant triples. Grey dashed arrow represents inheritance between the relations and "functionality".
  • Figure 4: NAL formalization of type III path
  • Figure 5: An illustration of type III path. The upper half represents positive version of the path and the lower half represents negative version. The dark cross represents the absence of the triple.
  • ...and 3 more figures