Naive Bayes (MultinomialNB)¶
Spam Detection¶
- The SMS Spam Collection is a set of SMS tagged messages that have been collected for SMS Spam research. It contains one set of SMS messages in English of 5,574 messages, tagged acording being ham (legitimate) or spam.
- The files contain one message per line. Each line is composed by two columns: v1 contains the label (ham or spam) and v2 contains the raw text.Importing the Relevant Libraries¶
In [1]:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
sns.set()
Importing the Dataset¶
In [2]:
url = "https://datascienceschools.github.io/Machine_Learning/Classification_Models_CaseStudies/emails.csv"
df = pd.read_csv(url)
df.head()
Out[2]:
Number & Percentage of Spam/Ham¶
In [3]:
spam = df[df['spam'] == 1]
ham = df[df['spam'] == 0]
print("Total Emails =", len(df))
print("\nSpams =", len(spam))
print("Percentage of Spams = {:.2f} %".format(1.*len(spam)/len(df)*100.0))
print("\nhams =", len(ham))
print("Percentage of Hams = {:.2f} %".format(1.*len(ham)/len(df)*100.0))
Countplot (Spam/Ham)¶
In [4]:
sns.countplot(df['spam'], palette= 'Set1')
plt.show()
Converting text to a matrix of token counts¶
In [5]:
from sklearn.feature_extraction.text import CountVectorizer
vectorizer = CountVectorizer()
text_verctorized = vectorizer.fit_transform(df['text'])
text = pd.DataFrame(text_verctorized.toarray())
text.head()
Out[5]:
Concatenatig df & text¶
In [6]:
df = df.drop('text', axis=1)
df = pd.concat([text, df], axis=1)
df.head()
Out[6]:
Array mapping from feature integer indices to feature name¶
- print(vectorizer.get_feature_names())Declaring the Dependent & the Independent Variables¶
In [7]:
X = df.iloc[:, :-1].values
y = df.iloc[:, -1].values
Splitting the Dataset into the Training Set and Test Set¶
In [8]:
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.20, random_state = 3)
Feature Scaling¶
- No need to scale dataTraining the Naive Bayes Model (MultinomialNB)¶
In [9]:
from sklearn.naive_bayes import MultinomialNB
model = MultinomialNB()
model.fit(X_train, y_train)
Out[9]:
Predicting the Test Set Results¶
In [10]:
y_pred = model.predict(X_test)
Confusion Matrix¶
In [11]:
from sklearn.metrics import confusion_matrix, accuracy_score
cm = confusion_matrix(y_test, y_pred)
accuracy = accuracy_score(y_test, y_pred)
print("Accuracy is: {:.2f} %".format(accuracy*100))
sns.heatmap(cm, annot=True, fmt='d')
plt.show()
Classification Report¶
In [12]:
from sklearn.metrics import classification_report
print(classification_report(y_test, y_pred))
K-Fold Cross Validation¶
In [13]:
from sklearn.model_selection import cross_val_score
accuracies = cross_val_score(estimator = model, X = X_train, y = y_train, cv = 10)
print("Accuracy: {:.2f} %".format(accuracies.mean()*100))
print("Standard Deviation: {:.2f} %".format(accuracies.std()*100))