Question

Spam Filtering using Naïve Bayesian classifier

In Spam filtering problem the examples or observations are the emails, and the features are the words in the email. When using words as features we suppose that some words are more likely to appear in spam than in non-spam, which is the basic premise underlying most spam filters.

The attached folder contains three files:

1. Spmabase.Documentation: has all the information about the dataset
   1. Source
   2. Number of examples (instances or observations)
   3. Number of attributes
   4. Attributes information
2. Spambase.names: has the names of the features, and the target
3. Spambase.data: contains the dataset

Write a Python code to:

1. train a Bayesian classifier to classify emails to spam email or non-spam email.
   1. Divide the dataset to a training set and a testing set.
   2. Train the classifier on the training set, and test the classifier on the testing set,
   3. use 70% of the dataset as a training and 30% of the data as testing.
2. Calculate the accuracy of the classifier.

The following documents can be found here  https://archive.ics.uci.edu/ml/machine-learning-databases/spambase/

• spambase.DOCUMENTATION
• spambase.data
• spambase.names

Thank you

Answer 1

Below is a screen shot of the python program to check indentation. Comments are given on every line explaining the code.

Below is the output of the program:


Below is the code to copy:

#CODE STARTS HERE----------------

import pandas as pd
from sklearn import metrics
from sklearn.naive_bayes import GaussianNB
from sklearn.model_selection import train_test_split

#Download spambase.data into the same directory as your code
#Add column names as integers from 0-57
df = pd.read_csv("spambase.data",header=None,names=range(58))
model = GaussianNB() #Load data into a dataframe
X= df.loc[:,:56] #Separate X from data
y= df.loc[:,57] #Separate y values(class) from data

#Split data into train(70%) and test(30%)
X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.3)
model.fit(X_train,y_train) #Train the model

pred = model.predict(X_test) #Predict output on test data
#Calculate and print accuracy
print("Accuracy: ",round(metrics.accuracy_score(y_test,pred),4))
#CODE ENDS HERE-----------------