Air quality prediction using neuro-fuzzy tools

A unified approach for integrating explicit and implicit knowledge in connectionist expert systems is proposed for air quality prediction. The explicit knowledge is represented by discrete fuzzy rules, which are directly mapped into an equivalent neural structure. Learning data set is incorporated in a neuro-fuzzy module, representing implicit knowledge. The combination of explicit and implicit knowledge modules is viewed as a fuzzy rule-based model of the problem and is implemented by a supervised trained gating network. Results are encouraging and show that the method is worthy of further research. Copyright (C) 2001 IFAC.