Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems

The generative approach to social science, in which agent-based simulations (or other complex systems models) are executed to reproduce a known social phenomenon, is an important tool for realist explanation. However, a generative model, when suitably calibrated and validated using empirical data, r...

Full description

Saved in:
Bibliographic Details
Main Authors: Tuong M. Vu, Charlotte Buckley, Hao Bai, Alexandra Nielsen, Charlotte Probst, Alan Brennan, Paul Shuper, Mark Strong, Robin C. Purshouse
Format: Article
Language:English
Published: Wiley 2020-01-01
Series:Complexity
Online Access:http://dx.doi.org/10.1155/2020/8923197
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850233341043802112
author Tuong M. Vu
Charlotte Buckley
Hao Bai
Alexandra Nielsen
Charlotte Probst
Alan Brennan
Paul Shuper
Mark Strong
Robin C. Purshouse
author_facet Tuong M. Vu
Charlotte Buckley
Hao Bai
Alexandra Nielsen
Charlotte Probst
Alan Brennan
Paul Shuper
Mark Strong
Robin C. Purshouse
author_sort Tuong M. Vu
collection DOAJ
description The generative approach to social science, in which agent-based simulations (or other complex systems models) are executed to reproduce a known social phenomenon, is an important tool for realist explanation. However, a generative model, when suitably calibrated and validated using empirical data, represents just one viable candidate set of entities and mechanisms. The model only partially addresses the needs of an abductive reasoning process—specifically it does not provide insight into other viable sets of entities or mechanisms nor suggests which of these are fundamentally constitutive for the phenomenon to exist. In this paper, we propose a new model discovery framework that more fully captures the needs of realist explanation. The framework exploits the implicit ontology of an existing human-built generative model to propose and test a plurality of new candidate model structures. Genetic programming is used to automate this search process. A multiobjective approach is used, which enables multiple perspectives on the value of any particular generative model—such as goodness of fit, parsimony, and interpretability—to be represented simultaneously. We demonstrate this new framework using a complex systems modeling case study of change and stasis in societal alcohol use patterns in the US over the period 1980–2010. The framework is successful in identifying three competing explanations of these alcohol use patterns, using novel integrations of social role theory not previously considered by the human modeler. Practitioners in complex systems modeling should use model discovery to improve the explanatory utility of the generative approach to realist social science.
format Article
id doaj-art-034a14e9c3b44f8692b4a71b1b93717d
institution OA Journals
issn 1076-2787
1099-0526
language English
publishDate 2020-01-01
publisher Wiley
record_format Article
series Complexity
spelling doaj-art-034a14e9c3b44f8692b4a71b1b93717d2025-08-20T02:02:57ZengWileyComplexity1076-27871099-05262020-01-01202010.1155/2020/89231978923197Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social SystemsTuong M. Vu0Charlotte Buckley1Hao Bai2Alexandra Nielsen3Charlotte Probst4Alan Brennan5Paul Shuper6Mark Strong7Robin C. Purshouse8School of Health and Related Research, University of Sheffield, Sheffield, UKAutomatic Control and Systems Engineering, University of Sheffield, Sheffield, UKAutomatic Control and Systems Engineering, University of Sheffield, Sheffield, UKAlcohol Research Group, Public Health Institute, Emeryville, CA, USAInstitute for Mental Health Policy Research, Centre for Addiction and Mental Health, Toronto, CanadaSchool of Health and Related Research, University of Sheffield, Sheffield, UKInstitute for Mental Health Policy Research, Centre for Addiction and Mental Health, Toronto, CanadaSchool of Health and Related Research, University of Sheffield, Sheffield, UKAutomatic Control and Systems Engineering, University of Sheffield, Sheffield, UKThe generative approach to social science, in which agent-based simulations (or other complex systems models) are executed to reproduce a known social phenomenon, is an important tool for realist explanation. However, a generative model, when suitably calibrated and validated using empirical data, represents just one viable candidate set of entities and mechanisms. The model only partially addresses the needs of an abductive reasoning process—specifically it does not provide insight into other viable sets of entities or mechanisms nor suggests which of these are fundamentally constitutive for the phenomenon to exist. In this paper, we propose a new model discovery framework that more fully captures the needs of realist explanation. The framework exploits the implicit ontology of an existing human-built generative model to propose and test a plurality of new candidate model structures. Genetic programming is used to automate this search process. A multiobjective approach is used, which enables multiple perspectives on the value of any particular generative model—such as goodness of fit, parsimony, and interpretability—to be represented simultaneously. We demonstrate this new framework using a complex systems modeling case study of change and stasis in societal alcohol use patterns in the US over the period 1980–2010. The framework is successful in identifying three competing explanations of these alcohol use patterns, using novel integrations of social role theory not previously considered by the human modeler. Practitioners in complex systems modeling should use model discovery to improve the explanatory utility of the generative approach to realist social science.http://dx.doi.org/10.1155/2020/8923197
spellingShingle Tuong M. Vu
Charlotte Buckley
Hao Bai
Alexandra Nielsen
Charlotte Probst
Alan Brennan
Paul Shuper
Mark Strong
Robin C. Purshouse
Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems
Complexity
title Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems
title_full Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems
title_fullStr Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems
title_full_unstemmed Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems
title_short Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems
title_sort multiobjective genetic programming can improve the explanatory capabilities of mechanism based models of social systems
url http://dx.doi.org/10.1155/2020/8923197
work_keys_str_mv AT tuongmvu multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT charlottebuckley multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT haobai multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT alexandranielsen multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT charlotteprobst multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT alanbrennan multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT paulshuper multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT markstrong multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems
AT robincpurshouse multiobjectivegeneticprogrammingcanimprovetheexplanatorycapabilitiesofmechanismbasedmodelsofsocialsystems