Combining Machine Learning and Ontology: A Systematic Literature Review

TL;DR

This systematic review analyzes 128 studies on integrating machine learning with ontologies, categorizing hybrid approaches and highlighting techniques that combine inductive and deductive reasoning in AI systems.

Contribution

It provides a comprehensive classification of hybrid AI methods combining machine learning and ontologies, including analysis of algorithms and comparison with recent approaches.

Findings

Identified three main categories: learning-enhanced ontologies, semantic data mining, and learning and reasoning systems.

Analyzed the machine learning algorithms used across studies.

Compared hybrid approaches with recent neuro-symbolic AI work.

Abstract

Motivated by the desire to explore the process of combining inductive and deductive reasoning, we conducted a systematic literature review of articles that investigate the integration of machine learning and ontologies. The objective was to identify diverse techniques that incorporate both inductive reasoning (performed by machine learning) and deductive reasoning (performed by ontologies) into artificial intelligence systems. Our review, which included the analysis of 128 studies, allowed us to identify three main categories of hybridization between machine learning and ontologies: learning-enhanced ontologies, semantic data mining, and learning and reasoning systems. We provide a comprehensive examination of all these categories, emphasizing the various machine learning algorithms utilized in the studies. Furthermore, we compared our classification with similar recent work in the field of hybrid AI and neuro-symbolic approaches.