Research@Lincoln
    • Login
     
    View Item 
    •   Research@Lincoln Home
    • Metadata-only (no full-text)
    • Metadata-only (no full-text)
    • View Item
    •   Research@Lincoln Home
    • Metadata-only (no full-text)
    • Metadata-only (no full-text)
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Mixed-method integration and advances in fuzzy cognitive maps for computational policy simulations for natural hazard mitigation

    Samarasinghe, Sandhya; Strickert, Graham E. H.
    Abstract
    Human systems need to be adaptive to the consequences of natural hazards. Public policy decisions on natural hazard mitigation can benefit from computational models that embody a comprehensive view of the system. Such models need to be transparent and integrate both expert and lay expert knowledge and experience in an efficient manner. By integrating hard and soft sciences within an overall systems framework, scientists, policy makers and communities can better understand how to improve adaptive capacity. We present a fuzzy cognitive map based Auto-Associative Neural Networks framework generated from a development mixed method integration (triangulation) for adaptive policy formulations. The specific policies relate to preparation for, response to, and recovery from earthquakes in mountainous ski-field environments – a case study chosen to highlight the framework. Three different data collection techniques – expert geomorphic assessments, semi-structured qualitative interviews with three stakeholder groups (experts and lay experts), and fuzzy cognitive maps (FCM) (node and arc maps of stakeholder perceptions) were employed. FCM were first analysed using Graph theory indices to determine map structure. Special attention was paid to subsequent processing of fuzzy cognitive maps (e.g., condensation and aggregation) with qualitative followed by quantitative means to simplify the FCM from the original total of 300 variables to 5 high-level themes to improve the efficacy of subsequent policy simulations. Specifically, the use of Self Organising Maps (SOM) to group concepts (condensation) and individual stakeholders (aggregation) into social group FCMs is a novel contribution to advancing FCM. In the process, SOM also enabled the embedment of nonlinear relationships inherent in the system in the simplified FCM allowing a platform for realistic and meaningful policy simulations based on collective perceptions. Specifically, each of the three simplified stakeholder group FCM and a total social group FCM was represented by Auto-Associative Neural Networks (AANN) which converts an FCM into a dynamical system that allows policy scenario simulations based on input from both expert and lay expert stakeholders. A policy scenario is the level of importance given to a set of concepts and their effects on the system behaviour as revealed by the simulations. We present the results from one of several policy simulations to highlight the effectiveness of the mixed-method integration leading to simplified-FCM based ANNN simulations. Results revealed the similarities and differences between stakeholder group responses in relation to the scenario analysed and how these formed collective responses in the total social group map. Furthermore, outcomes of group and total social group simulations could be interpreted from individual and group stakeholder FCMs giving credibility to the mixed-method approach.... [Show full abstract]
    Keywords
    policy simulation; fuzzy cognitive maps; auto-associative neural networks; self-organizing maps; natural hazard mitigation; earthquakes; mixed-method triangulation; Environmental Engineering
    Fields of Research
    080108 Neural, Evolutionary and Fuzzy Computation; 170205 Neurocognitive Patterns and Neural Networks; 040604 Natural Hazards; 040601 Geomorphology and Regolith and Landscape Evolution
    Date
    2013-01
    Type
    Journal Article
    Collections
    • Metadata-only (no full-text) [4843]
    View/Open
    Share this

    on Twitter on Facebook on LinkedIn on Reddit on Tumblr by Email

    DOI
    https://doi.org/10.1016/j.envsoft.2012.06.008
    Metadata
     Expand record
    © 2012 Elsevier Ltd. All rights reserved
    Citation
    Samarasinghe, S., & Strickert, G. (2013). Mixed-method integration and advances in fuzzy cognitive maps for computational policy simulations for natural hazard mitigation. Environmental Modelling and Software, 39, 188-200. doi:10.1016/j.envsoft.2012.06.008
    This service is maintained by Learning, Teaching and Library
    • Archive Policy
    • Copyright and Reuse
    • Deposit Guidelines and FAQ
    • Contact Us
     

     

    Browse

    All of Research@LincolnCommunities & CollectionsTitlesAuthorsKeywordsBy Issue DateThis CollectionTitlesAuthorsKeywordsBy Issue Date

    My Account

    LoginRegister

    Statistics

    View Usage Statistics
    This service is maintained by Learning, Teaching and Library
    • Archive Policy
    • Copyright and Reuse
    • Deposit Guidelines and FAQ
    • Contact Us