A Boxology of Design Patterns for Hybrid Learning and Reasoning Systems

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

We propose a set of compositional design patterns to describe a large variety of systems that combine statistical techniques from machine learning with symbolic techniques from knowledge representation. As in other areas of computer science (knowledge engineering, software engineering, ontology engineering, process mining and others), such design patterns help to systematize the literature, clarify which combinations of techniques serve which purposes, and encourage re-use of software components. We have validated our set of compositional design patterns against a large body of recent literature.
Original languageEnglish
Article number3
Pages (from-to)97-124
Number of pages28
JournalJOURNAL OF WEB ENGINEERING
Volume18
Issue number1-3
DOIs
Publication statusPublished - Jan 2019

Bibliographical note

Abstract in CEUR vol. 2491.

Keywords

  • Hybrid systems
  • neurosymbolic systems
  • knowledge representation
  • machine learning
  • design patters

Cite this