A new fuzzy-based methodology for modelling risk predictions

A new paper, presented by Nasser Amaitlik of Aston University at the 2017 Computing Conference in London, proposes a new fuzzy-based methodology which can be applied to the modelling of mental-health risk predictions.

The paper proposes a hierarchical fuzzy rule-based model that uses a simplified and effective method for supporting the elicitation of the fuzzy rules and adapting uncertainty propagation that can be intuitively understood by human experts. The model is initially applied to predicting the risk of failure in water pipe systems, but Nasser demonstrates how it may be extended to other knowledge domains based on human expertise, such as mental-health risk assessment.

Adapting this model to aid in improving assessment of risk in patients with mental health difficulties is the next step in this research, and is closely related to the work currently done on GRaCE.

View the full paper, or an interview with Nasser.