Neural, symbolic and neural-symbolic reasoning on knowledge graphs

TLDR

Knowledge‑graph reasoning underpins machine‑learning tasks such as extraction, retrieval, and recommendation, yet symbolic methods struggle with noisy data while neural approaches offer robustness at the cost of interpretability, prompting hybrid strategies. This survey reviews the evolution of symbolic, neural, and hybrid reasoning on knowledge graphs and outlines future research directions. The authors analyze knowledge‑graph completion and question‑answering tasks within a unified reasoning framework.

Abstract

Knowledge graph reasoning is the fundamental component to support machine learning applications such as information extraction, information retrieval, and recommendation. Since knowledge graphs can be viewed as the discrete symbolic representations of knowledge, reasoning on knowledge graphs can naturally leverage the symbolic techniques. However, symbolic reasoning is intolerant of the ambiguous and noisy data. On the contrary, the recent advances of deep learning have promoted neural reasoning on knowledge graphs, which is robust to the ambiguous and noisy data, but lacks interpretability compared to symbolic reasoning. Considering the advantages and disadvantages of both methodologies, recent efforts have been made on combining the two reasoning methods. In this survey, we take a thorough look at the development of the symbolic, neural and hybrid reasoning on knowledge graphs. We survey two specific reasoning tasks — knowledge graph completion and question answering on knowledge graphs, and explain them in a unified reasoning framework. We also briefly discuss the future directions for knowledge graph reasoning.