Skip to content

Latest commit

 

History

History
9 lines (4 loc) · 718 Bytes

README.md

File metadata and controls

9 lines (4 loc) · 718 Bytes

In this project the aim was to predict that if a person falls under the category of normal weight or the person is obese. There are 7 categories of this. Normal, Overweight I, Overweight II, Obesetype I, II, III and Underweight.

As we can see that this a problem of multiclassification, for this project we have used KNN, SVM, Decision Tree and Random Forest Algorithms. This data contains both numerical and categorical data. They were transfomed individually.

The metrics - the confusion matrix, accuracy, precision, recall, f1 score, micro and macro average scores were studied closely and it was observed that Random forest does the best. To find the best parameters, hyperparameter tunigs are also done.