Lab Activities $(5$ Marks$)$

First Activity $($Students Grade Class Classification based KNN$)$

The dataset of students' performance contains $2392$ instances of data, and each instance is described by $7$ attributes $($including the label attribute$).$ Students are classified to five grades from $0$ to $4.$

Activity Requirements:

Deploy a code for each of the following parts $($each part in a separate file$)$ with the following activities to detect students' grade class using the previous mentioned dataset $($Students Performance.csv$)$

Load the dataset and present the characteristics of the dataset

Build a $k-$nearest neighbor classification model with test size $20\%$ and $k=10$

Assess the performance of the classifier using Training and test $($compute accuracy and cross validation$).$

Using cross$-$validation, compute the best value of the parameter $k($with a range of $k=1,3,5,$dots,$30)$

Plot a relation between k and the misclassification error

Compute the accuracy of the model using the best k $.$

Notes:

The best value of $k$ ensures the lowest misclassification error.

The misclassification error $=1-$ classification score

Second Activity $($Classification based Na$$ve Bayes$)$

The dataset of customers' segmentation contains $7741$ instances of data, and each instance is described by $6$ attributes $($including the label attribute$).$ Students are classified to four segments grades from A to D$.$

Activity Requirements:

Deploy a code for each of the following parts $($each part in a separate file$)$ with the following activities to detect customers' segment class using the previous mentioned dataset $($Customers segmentation.csv$)$

Load the dataset and present the characteristics of the dataset

Split the dataset into features $(x)$ and labels $($ y $)$

Convert categorical variables into a numerical representation

Split the data into training and testing sets

Create a Naive Bayes classifier

Train the classifier

Make predictions on the test set

Calculate the accuracy of the classifier

Calculate conditional probabilities