Fig 1.
The methodological flowchart of the study.
Table 1.
Bivariate analysis between ever-married women’s nutritional outcomes, socioeconomic, health, and demographic characteristics from BDHS 2017–18.
Table 2.
Specifications of SML algorithms.
Table 3.
Performance of different regression algorithms in BMI prediction using 10-fold cross-validation.
Fig 2.
Performance evaluation of different classification algorithms in predicting nutritional status using 10-fold cross-validation.
Here, LTR = Logistic regression; LLTR = LASSO logistic regression; DT = Decision tree; RF = Random forest; CT = Conditional inference tree; ADB = Adaptive boosting; XGB = eXtreme Gradient Boosting; KNN = K- nearest neighbor; SVM = Support vector machine; NN = Neural network; NB = Naïve Bayes.
Fig 3.
Performance evaluation of different classification algorithms in predicting nutritional status based on Asian cutoff using 10-fold cross-validation.
Here, LTR = Logistic regression; LLTR = LASSO logistic regression; DT = Decision tree; RF = Random forest; CT = Conditional inference tree; ADB = Adaptive boosting; XGB = eXtreme Gradient Boosting; KNN = K- nearest neighbor; SVM = Support vector machine; NN = Neural network; NB = Naïve Bayes.
Table 4.
Feature importance ranking from the SML algorithms.
Fig 4.
Conditional inference tree for BMI prediction using 10-fold cross-validation.
Each terminal node of this algorithm provides the average predicted value, the number of observations used, and the sum of squared error. NE, PE, SE, and HE represent no education, primary education, secondary education, and higher degree (such as college), respectively.
Fig 5.
Conditional inference decision tree for underweight using 10-fold cross-validation.
Each terminal node of this algorithm provides the underweight status (Yes or No), the number of observations used, and the classification error.
Fig 6.
Conditional inference decision tree for overweight using 10-fold cross-validation.
The categories of the education variable are the same as those of partner’s education. Each terminal node of this algorithm provides the overweight status (Yes or No), the number of observations used, and the classification error.
Fig 7.
Conditional inference decision tree for obesity using 10-fold cross-validation.
On the partner’s employment variable, BC, WC, and UNEMP respectively represent blue-collar job, white-collar-job, and unemployed. Each terminal node of this algorithm provides the obesity status (Yes or No), the number of observations used, and the classification error.