In this problem you are required to apply various classification techniques on a benchmark dataset: Diabetes Risk Prediction given as a CSV file in the attachment $($source: Kaggle$).$

This dataset is a collection of medical and demographic data from patients, along with their diabetes status $($positive or negative$).$ The data includes features such as age, gender, polyuria, polydipsia, sudden weight loss, weakness, polyphagia, visual blurring, itching, irritability, delayed healing etc. This dataset can be used to build machine learning models to predict diabetes in patients based on their medical history and demographic information.

$1$

a$.$ Apply a $75-25$ random split into train and test data and then apply the classification algorithms discussed in our classes:

i$.$Decision trees

ii$.$KNN

iii.Support Vector Machines

iv$.$Logistic Regression

v$.$Na$$ve Bayes

For each classification technique that is applied, the overall performance report should be printed. Briefly compare the results.

b$.$ Furthermore you are going to apply a simplified version of the k$-$fold cross validation method. The simplified $k-$fold cross validation method proceeds in this way:

Shuffle the dataset

Divide the dataset into $k$ equal partitions

For each of the partitions:

Apply the classification model using the union of $($k$-1)$ other partitions as training set

Test it on the current partition

Analyze the performance $($precision$,$ recall, accuracy$)$ for this case

Generate the overall performance report, taking the averages of the found results.

Apply a $5-$fold cross validation methodology, as described above, for all the classification techniques that we have studied in our course:

ii$.$Decision trees

iii.KNN

iv$.$Support Vector Machines

v$.$Logistic Regression

vi$.$Na$$ve Bayes

For each classification technique that is applied, the overall performance report should be printed. Briefly compare the results.