Diarrhea is one of the most common infectious diseases that affect people of all ages and is a serious public health concern around the world. The main causes of diarrhea include food quality, water, indoor meteorological, and outdoor meteorological conditions. In this study, a dew computing-assisted smart monitoring framework is developed to evaluate the relationship among the health, indoor meteorological, and food factors of an individual to predict the cause of diarrhea with the scale of severity. Smart sensors are utilized at the physical layer to collect the targeted parameters of health, indoor meteorological, and food of the individual. The captured events are classified at the cyber layer by utilizing the Probabilistic Weighted-Naïve Bayes (PW-NB) classification approach for quantifying abnormal health events. Furthermore, a Multi-scale Gated Recurrent Unit (M-GRU) is suggested to obtain the scale of severity by analyzing the correlation between irregular health, food, and environmental events. In this manner, the proposed model M-GRU has achieved a high precision value of (93.26%\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$93.26\%$$\end{document}), whereas, LSTM, RNN, SVM achieved the precision value of (89.13%\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$89.13\%$$\end{document}), (90.43%\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$90.43\%$$\end{document}), (88.23%\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$88.23\%$$\end{document}), respectively. In addition, the precision value of the PW-NB is (97.15%\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$97.15\%$$\end{document}), which is also higher as compared to KNN (93.25%\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$93.25\%$$\end{document}) and DT (96.91%\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$96.91\%$$\end{document}). The outcome of the proposed solutions is shown the higher Precision values on dew computing and cloud computing. Moreover, a comparative analysis defines the prediction effectiveness of the proposed solution over several other decision-making solutions with regards to event classification, severity determination, monitoring stability, and prediction efficiency.
