This is a data exploring study. It can be regarded as a work performed for data mining. You can see the descriptions along with the codes to be helpful in this regard. The data set used is taken from Sklearn library.
The wine dateset is a classic and very easy multi-class classification dataset.
Classes 3
Samples per class [59,71,48]
Samples total 178
Dimensionality 13
Features real, positive
This assignment is suggested to complete on Google colab to benefit from its GPU support.
Linear SVM, NonLinear SVM, MLP classifiers have been applied in general, and some results are as follows; the detailed form is in the outputs in the code.
Comparing of prediction results of each classifier based on accuracy
SVC with linear : 0.5925925925925926
SVC with RBF : 0.7407407407407407
SVC with polynomial : 0.7407407407407407
SVC with Sigmoid : 0.35185185185185186
You can also check the addresses below to get more information and use as a resource.
[1] https://jonathonbechtel.com/blog/2018/02/06/wines/
[2] https://www.kaggle.com/mfirat/red-wine-quality-classification-with-kernel-svm
[3] https://medium.com/analytics-vidhya/step-by-step-guide-for-predicting-wine-quality-using-scikit-learn-de5869f8f91a