Neural, symbolic and neural-symbolic reasoning on knowledge graphs

By Jing Zhang, Bo Chen, Lingxi Zhang, Xirui Ke, Haipeng Ding. 2021

149 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Natural Language Processing, Graph Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Nan Duan
2. Ming Zhou
3. Weizhu Chen
4. Qian Liu
5. Jian-Guang Lou
6. Zeqi Lin
7. Qiang Fu
8. Yue Liu

In Natural Language Processing

1. Nan Duan
2. Ming Zhou
3. Linhao Luo
4. Qian Liu
5. Yan Gao
6. Zhumin Chen
7. Xindong Wu
8. Morteza Ziyadi

In Graph Learning

1. Xinwang Liu
2. Shirui Pan
3. Nan Duan
4. Wanjun Zhong

Citations in the corpus

• Reasoning Like Program Executors (2022, background): “, neuro-symbolic reasoning, has been a major area of interest (Besold et al., 2017; Zhang et al., 2021).”
• Unifying Large Language Models and Knowledge Graphs: A Roadmap (2023, background): “Besides, they are renowned for their symbolic reasoning ability [22], which generates interpretable results.”
• From LSAT: The Progress and Challenges of Complex Reasoning (2021, methodology): “Advanced methods to solve complex reasoning problems, no matter what specific reasoning skill is required, can be summarized as following three types: symbolic models, neural models and neural-symbolic models [15, 16, 17].”
• A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal (2022, methodology): “Besides, [15] divides KGR models into three types, i.e., symbolic, neural, and hybrid.”