A serious game to parameterize Bayesian networks: Validation in a case study in northeastern Madagascar. (December 2019)

• Record Type: Journal Article
• Authors: Celio, Enrico; Andriatsitohaina, R. Ntsiva N.; Zaehringer, Julie G.
• Is Part Of: Environmental modelling & software. Volume 122(2019)
• Publication Date: 2019-12
• Subjects: Bayesian network -- (Serious) game -- Participation -- Validation -- Parameterization; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2019.104525
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: The management of multi-use landscapes is challenging, but essential when aiming at preserving the potential for ecosystem service provision. Land-use decisions lay at the center of this challenge. While land-use decision models may help to transparently grasp land-use decisions, the parameterization of such models is difficult as human decision-making is often not rational. We show here how we used a serious game to parameterize a Bayesian network-based land-use decision model. To elicit validation, game outputs are transformed to conditional probabilities and compared to conditional probabilities parameterized via a questionnaire and workshop exercises. The analysis of four types of validity shows encouraging results for criterion, respondent-related and practice-related validation. However, content validation (sensitivity analysis) was disappointing initially. We discuss how the success in validation quality may be related to the design of the game and conclude that the transfer from a game to Bayesian networks could improve the parameterization quality. Highlights: We parameterize a Bayesian network node by using a game. We validate the results in four validity perspectives. Three out of four validity perspectives show encouraging results. Insights gained via the game revealed usually intangible knowledge.