An Interval Type-2 Fuzzy LSTM Algorithm for Modeling Environmental Time-Series Prediction

The statistical attributes of the non-stationary problems such as air quality and other natural phenomena frequently changed. Type-2 fuzzy logic is a robust and capable model to cope with high-order uncertainties associated with non-stationary time-dependent features. This research's main objective is to present a novel Fuzzy Deep LSTM (IT2FLSTM) model to predict air quality for Tehran and Beijing in a short and long time series scale. The proposed model has been evaluated on a real dataset that contains the one-decade information about outdoor pollutants from April 2011 to November 2020 in Tehran and Beijing. The IT2FLSTM model was evaluated using a ROC curve analysis and validated using 10-fold cross-validation. The results confirm the IT2FLSTM model's superiority with an average area under the ROC curve (AUC) of 97 % and a 95% confidence interval of [95-98] %. The proposed IT2FLSTM model promises to predict complex problems to make strategic prevention decisions to save more lives.