Table 1 Comparison of literature review of the very popular and highly cited work.
From: Machine learning-based predictive modelling for the enhancement of wine quality
Study | ML technique | Wine characteristics considered | Prediction of wine quality | Results | Drawbacks |
|---|---|---|---|---|---|
Cortez et al.8 | Regression | Physicochemical data | Continuous scale from 0 to 10 | High prediction accuracy | N/A |
Agarwal et al.9 | Deep learning | Limited data set and features | N/A | N/A | Limited data set and available features |
Aich et al.10 | PCA, RFE, Nonlinear decision tree-based classifiers | N/A | Crucial elements for choosing high-quality wines | N/A | N/A |
Mahima et al.16 | Random Forest, KNN | N/A | Good, Average, or Bad | Improved prediction accuracy with KNN | N/A |
Kumar et al.12 | RF, SVM, NB | N/A | N/A | Comparison of performance metrics between training and testing sets | N/A |
Shaw et al.13 | SVM, RF, Multilayer perceptron | N/A | Accuracy comparison | RF produces best results with accuracy of 81.96% | N/A |
Bhardwaj et al.14 | Four feature selection techniques, seven machine learning algorithms | Chemical and physicochemical data | Significant attributes for predicting wine quality | N/A | N/A |
Tiwari et al.15 | Mathematical model, ML techniques | Sensory and chemical data | Wine quality | N/A | N/A |
Ma et al.17 | Deep learning | Physicochemical data | Wine type | High prediction accuracy | N/A |
Prez et al.18 | PCA, SVM | Physicochemical data | Wine quality | N/A | N/A |
Gupta et al.11 | RF, SVM, NB | Sensory data | Wine quality | N/A | N/A |
