In Python, Naive Bayes algorithm use GaussianNBClassifier to apply Naive Bayes algorithm in Python.

In this step, we are using StudentEvent_Resample.xlsx dataset. The value for this dataset has been standardized in Rapidminer and its contains 100 rows with 11 columns.

#Naive Bayes Model

data_nb = df[['Assignment','Forum','Activity','LectureNote',

'Tutorial','Questionnaire','Quiz','MarksBin']].copy()

data_nb.info()

#Decision Tree Model

#import the classifier

from sklearn import metrics

from sklearn.tree import DecisionTreeClassifier

from sklearn.metrics import accuracy_score

from sklearn.metrics import classification_report

from sklearn.metrics import confusion_matrix

import warnings

warnings.filterwarnings('ignore')

# Split into predictor and response dataframes.

X_nb = data_nb.drop('MarksBin', axis=1)

y_nb = data_nb['MarksBin']

X_nb.shape,y_nb.shape

from sklearn.model_selection import train_test_split

X_train_nb, X_test_nb, y_train_nb, y_test_nb = train_test_split(X_nb, y_nb, test_size = 0.30, random_state = 30)

X_train_nb1, X_test_nb1, y_train_nb1, y_test_nb1 = train_test_split(X_nb, y_nb, test_size = 0.50, random_state = 30)

from sklearn.naive_bayes import GaussianNB

#lets try with another classifier: Naive Bayes (70:30)

clf_nb = GaussianNB()

clf_nb.fit(X_train_nb, y_train_nb)

y_pred_nb = clf_nb.predict(X_test_nb)

print("Performance Accuracy: {:.2f} %".format(metrics.accuracy_score(y_pred_nb,y_test_nb)*100))

print("Performance Error: {:.2f} %".format(100-(metrics.accuracy_score(y_test_nb,y_pred_nb)*100)))

print("Classification Report")

classify_nb = metrics.classification_report(y_pred_nb,y_test_nb);

print(classify_nb)

print("Confusion Matrix")

confusion_matrix_nb = pd.crosstab(y_test_nb, y_pred_nb, rownames=['Actual'], colnames=['Predicted'], margins = True)

print (confusion_matrix_nb)

from sklearn.naive_bayes import GaussianNB

#lets try with another classifier: Naive Bayes (50:50)

clf_nb1 = GaussianNB()

clf_nb1.fit(X_train_nb1, y_train_nb1)

y_pred_nb1 = clf_nb.predict(X_test_nb1)

print("Performance Accuracy: {:.2f} %".format(metrics.accuracy_score(y_pred_nb1,y_test_nb1)*100))

print("Performance Error: {:.2f} %".format(100-(metrics.accuracy_score(y_test_nb1,y_pred_nb1)*100)))

print("Classification Report")

classify_nb1 = metrics.classification_report(y_pred_nb1,y_test_nb1);

print(classify_nb1)

print("Confusion Matrix")

confusion_matrix_nb1 = pd.crosstab(y_test_nb1, y_pred_nb1, rownames=['Actual'], colnames=['Predicted'], margins = True)

print (confusion_matrix_nb1)

#hyperparameter tuning for Naive Bayes Model

import numpy as np

from sklearn.model_selection import train_test_split,GridSearchCV

import matplotlib.pyplot as plt

import seaborn as sns

from sklearn.model_selection import cross_val_score

from sklearn.metrics import accuracy_score, confusion_matrix

sns.set_style("whitegrid")

np.logspace(0,-9, num=10)

from sklearn.model_selection import RepeatedStratifiedKFold

cv_method = RepeatedStratifiedKFold(n_splits=5, n_repeats=3, random_state=999)

from sklearn.preprocessing import PowerTransformer

params_NB = {'var_smoothing': np.logspace(0,-9, num=100)}

gs_NB = GridSearchCV(estimator=clf_nb, param_grid=params_NB, cv=cv_method,verbose=1,scoring='accuracy')

Data_transformed = PowerTransformer().fit_transform(X_test_nb)

gs_NB.fit(Data_transformed, y_test_nb);

results_NB = pd.DataFrame(gs_NB.cv_results_['params'])

results_NB['test_score'] = gs_NB.cv_results_['mean_test_score']

# predict the target on the test dataset

predict_test_nb = gs_NB.predict(Data_transformed)

# Accuracy Score on test dataset

accuracy_test_nb = accuracy_score(y_test_nb,predict_test_nb)

print('Accuracy_score on test dataset : ', (accuracy_test_nb*100))

gs_NB1 = GridSearchCV(estimator=clf_nb, param_grid=params_NB, cv=cv_method,verbose=1,scoring='accuracy')

Data_transformed1 = PowerTransformer().fit_transform(X_test_nb1)

gs_NB1.fit(Data_transformed1, y_test_nb1);

results_NB1 = pd.DataFrame(gs_NB1.cv_results_['params'])

results_NB1['test_score'] = gs_NB1.cv_results_['mean_test_score']

# predict the target on the test dataset

predict_test_nb1 = gs_NB1.predict(Data_transformed1)

# Accuracy Score on test dataset

accuracy_test_nb1 = accuracy_score(y_test_nb1,predict_test_nb1)

print('Accuracy_score on test dataset : ', (accuracy_test_nb1*100))

This is how we apply the suggestion value into our model.

#after tuning test on 30:70 ratio

X_test_nb = Data_transformed

y_pred_nb = predict_test_nb

print("\nPerformance Accuracy: {:.2f} %".format(metrics.accuracy_score(y_pred_nb,y_test_nb)*100))

print("Performance Error: {:.2f} %".format(100-(metrics.accuracy_score(y_test_nb,y_pred_nb)*100)))

print("\nClassification Report")

classify_nb2 = metrics.classification_report(y_pred_nb,y_test_nb);

print(classify_nb2)

print("\nConfusion Matrix")

confusion_matrix_nb2 = pd.crosstab(y_test_nb, y_pred_nb, rownames=['Actual'], colnames=['Predicted'], margins = True)

print (confusion_matrix_nb2)

#after tuning test on 50:50 ratio

X_test_nb1 = Data_transformed1

y_pred_nb1 = predict_test_nb1

print("\nPerformance Accuracy: {:.2f} %".format(metrics.accuracy_score(y_pred_nb1,y_test_nb1)*100))

print("Performance Error: {:.2f} %".format(100-(metrics.accuracy_score(y_test_nb1,y_pred_nb1)*100)))

print("\nClassification Report")

classify_nb3 = metrics.classification_report(y_pred_nb1,y_test_nb1);

print(classify_nb3)

print("\nConfusion Matrix")

confusion_matrix_nb3 = pd.crosstab(y_test_nb1, y_pred_nb1, rownames=['Actual'], colnames=['Predicted'], margins = True)

print (confusion_matrix_nb3)

From the above results, we can see that how effective hyperparameter tuning for our Naive Bayes model. For the ratio of 30:70, the performance accuracy increase to 83.3%. For the ratio of 50:50, the performance accuracy increase to 84%. From all the results, we can conclude that the best ratio for the Naive Bayes model using Python is for the ratio of 50:50.