Using model replication to improve the reliability of agent-based models

Wei Zhong, Yushim Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

The basic presupposition of model replication activities for a computational model such as an agent-based model (ABM) is that, as a robust and reliable tool, it must be replicable in other computing settings. This assumption has recently gained attention in the community of artificial society and simulation due to the challenges of model verification and validation. Illustrating the replication of an ABM representing fraudulent behavior in a public service delivery system originally developed in the Java-based MASON toolkit for NetLogo by a different author, this paper exemplifies how model replication exercises provide unique opportunities for model verification and validation process. At the same time, it helps accumulate best practices and patterns of model replication and contributes to the agenda of developing a standard methodological protocol for agent-based social simulation.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages118-127
Number of pages10
Volume6007 LNCS
DOIs
StatePublished - 2010
Event3rd International Conference on Social Computing, Behavioral Modeling, and Prediction, SBP 2010 - Bethesda, MD, United States
Duration: Mar 30 2010Mar 31 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6007 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other3rd International Conference on Social Computing, Behavioral Modeling, and Prediction, SBP 2010
CountryUnited States
CityBethesda, MD
Period3/30/103/31/10

Fingerprint

Agent-based Model
Replication
Model Verification
Verification and Validation
Model Validation
Social Simulation
Public Services
Agent-based Simulation
Best Practice
Accumulate
Model
Computational Model
Exercise
Java
Computing
Simulation
Network protocols

Keywords

  • Agent-based model
  • Model replication
  • Validation
  • Verification

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Zhong, W., & Kim, Y. (2010). Using model replication to improve the reliability of agent-based models. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6007 LNCS, pp. 118-127). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6007 LNCS). https://doi.org/10.1007/978-3-642-12079-4_17

Using model replication to improve the reliability of agent-based models. / Zhong, Wei; Kim, Yushim.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6007 LNCS 2010. p. 118-127 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6007 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhong, W & Kim, Y 2010, Using model replication to improve the reliability of agent-based models. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 6007 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6007 LNCS, pp. 118-127, 3rd International Conference on Social Computing, Behavioral Modeling, and Prediction, SBP 2010, Bethesda, MD, United States, 3/30/10. https://doi.org/10.1007/978-3-642-12079-4_17
Zhong W, Kim Y. Using model replication to improve the reliability of agent-based models. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6007 LNCS. 2010. p. 118-127. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-12079-4_17
Zhong, Wei ; Kim, Yushim. / Using model replication to improve the reliability of agent-based models. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6007 LNCS 2010. pp. 118-127 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{253a77d2b5094828b39eca09bf6e2517,
title = "Using model replication to improve the reliability of agent-based models",
abstract = "The basic presupposition of model replication activities for a computational model such as an agent-based model (ABM) is that, as a robust and reliable tool, it must be replicable in other computing settings. This assumption has recently gained attention in the community of artificial society and simulation due to the challenges of model verification and validation. Illustrating the replication of an ABM representing fraudulent behavior in a public service delivery system originally developed in the Java-based MASON toolkit for NetLogo by a different author, this paper exemplifies how model replication exercises provide unique opportunities for model verification and validation process. At the same time, it helps accumulate best practices and patterns of model replication and contributes to the agenda of developing a standard methodological protocol for agent-based social simulation.",
keywords = "Agent-based model, Model replication, Validation, Verification",
author = "Wei Zhong and Yushim Kim",
year = "2010",
doi = "10.1007/978-3-642-12079-4_17",
language = "English (US)",
isbn = "3642120784",
volume = "6007 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "118--127",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Using model replication to improve the reliability of agent-based models

AU - Zhong, Wei

AU - Kim, Yushim

PY - 2010

Y1 - 2010

N2 - The basic presupposition of model replication activities for a computational model such as an agent-based model (ABM) is that, as a robust and reliable tool, it must be replicable in other computing settings. This assumption has recently gained attention in the community of artificial society and simulation due to the challenges of model verification and validation. Illustrating the replication of an ABM representing fraudulent behavior in a public service delivery system originally developed in the Java-based MASON toolkit for NetLogo by a different author, this paper exemplifies how model replication exercises provide unique opportunities for model verification and validation process. At the same time, it helps accumulate best practices and patterns of model replication and contributes to the agenda of developing a standard methodological protocol for agent-based social simulation.

AB - The basic presupposition of model replication activities for a computational model such as an agent-based model (ABM) is that, as a robust and reliable tool, it must be replicable in other computing settings. This assumption has recently gained attention in the community of artificial society and simulation due to the challenges of model verification and validation. Illustrating the replication of an ABM representing fraudulent behavior in a public service delivery system originally developed in the Java-based MASON toolkit for NetLogo by a different author, this paper exemplifies how model replication exercises provide unique opportunities for model verification and validation process. At the same time, it helps accumulate best practices and patterns of model replication and contributes to the agenda of developing a standard methodological protocol for agent-based social simulation.

KW - Agent-based model

KW - Model replication

KW - Validation

KW - Verification

UR - http://www.scopus.com/inward/record.url?scp=78650373803&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650373803&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-12079-4_17

DO - 10.1007/978-3-642-12079-4_17

M3 - Conference contribution

AN - SCOPUS:78650373803

SN - 3642120784

SN - 9783642120787

VL - 6007 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 118

EP - 127

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -