Inference reasoning on fishers' knowledge using Bayesian causal maps

Abstract

Scientists and managers are not the only holders of knowledge regarding environmental issues: other stakeholders such as farmers or fishers do have empirical and relevant knowledge. Thus, new approaches for knowledge representation in the case of multiple knowledge sources, but still enabling reasoning, are needed. Cognitive maps and Bayesian networks constitute some useful formalisms to address knowledge representations. Cognitive maps are powerful graphical models for knowledge gathering or displaying. If they offer an easy means to express individuals judgments, drawing inferences in cognitive maps remains a difficult task. Bayesian networks are widely used for decision making processes that face uncertain information or diagnosis. But they are difficult to elicitate. To take advantage of each formalism and to overcome their drawbacks, Bayesian causal maps have been developed. In this approach, cognitive maps are used to build the network and obtain conditional probability tables. We propose here a complete framework applied on a real problem. From the different views of a group of shellfish dredgers about their activity, we derive a decision facilitating tool, enabling scenarios testing for fisheries management.qualitative modelling, cognitive maps, bayesian networks, fisher's knowledge, fisheries management, qualitative decision support 1

Tailoring Cognitive Mapping Analysis Methods to Different Management Styles of Collective Action by Handling Actor Reasoning Diversity

“…This majority summary was used as a base for a simulation model generating inference reasoning to estimate how the fishers would perceive the evolution of their activity in the face of certain changes (environmental, policy, etc.) (in Bonneau de Beaufort et al, 2015). The cognitive map summary was used in the ICZM process, as a way to include the main considerations of the fishers’ group in the multi-actor dialogue.

Figure 1.Cognitive mapping method and analysis for design experiment 1“Considering actor reasoning in top-down decision-making (scallop fishers)”.

…”

“…Cognitive mapping was used to fuel an optimized model to generate a summary of the reasoning of this category of actors. Bonneau de Beaufort et al (2015), a researcher associated to the project, developed a decision-making tool to explore scenarios with a simulation model. The researchers defined and framed the problem to address.…”

“…Cognitive maps were automatically compared and aggregated, generating cognitive map summaries, with a process detailed in Bonneau de Beaufort et al (2015), who used the same data (for more detail on this aggregation process, see Supporting Information). Only one majority cognitive map summary was used within the ICZM project (Figure 1): it showed elements that had been cited by at least seven fishers, including the main diverging opinions among fishers (dissensus, indicated in red).…”

“…Although it is easy and quick to explain the principle of a cognitive map, not all actors feel comfortable with its formalism. Indeed, drawing cognitive maps requires certain visualization and information coding capacities which are not necessarily intuitive processes (Bonneau de Beaufort et al, 2015). Moreover, the information collected by the interviewer can be biased by his presence.…”

Risk Assessment of An Earthquake-Collapse-Landslide Disaster Chain by Bayesian Network and Newmark Models

“…The quantitative part of a BN is the network parameters, which specify the conditional probability of every node under the conditions of its father nodes [40]. The dependencies between parent nodes with child nodes are quantitatively represented by a conditional probability table (CPT), which can be obtained from expert decision [42,43] or from training based on historical data [44,45]. Each node of a BN model is divided into a finite suit of status values represented by qualitative or quantitative values.…”