Scribd
Upload
37 views2 pages

Exp-9 - Jupyter Notebook

The document discusses building a spam classifier using Naive Bayes. It loads and preprocesses a spam dataset, builds vocabularies of words for spam and ham emails, and calculates accuracy, precision and recall of the classifier on a test set.

Uploaded by

Dhana Lakshmi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
37 views2 pages

Exp-9 - Jupyter Notebook

The document discusses building a spam classifier using Naive Bayes. It loads and preprocesses a spam dataset, builds vocabularies of words for spam and ham emails, and calculates accuracy, precision and recall of the classifier on a test set.

Uploaded by

Dhana Lakshmi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

5/23/22, 8:08 PM Exp-9 - Jupyter Notebook

In [12]:

import numpy as np
import pandas as pd
from collections import Counter

In [13]:

data = pd.read_csv('./data/spam.csv')
data = data[['v1', 'v2']]
data.head()

Out[13]:

v1 v2

0 ham Go until jurong point, crazy.. Available only ...

1 ham Ok lar... Joking wif u oni...

2 spam Free entry in 2 a wkly comp to win FA Cup fina...

3 ham U dun say so early hor... U c already then say...

4 ham Nah I don't think he goes to usf, he lives aro...

In [14]:

data = data.sample(frac=1)
train, test = data[:4000], data[4000:]
X_train, X_test, y_train, y_test = train['v2'], test['v2'], train['v1'], test['v1']

In [15]:

X_train.shape, X_test.shape, y_train.shape, y_test.shape

Out[15]:

((4000,), (1572,), (4000,), (1572,))

In [16]:

vocab_spam = Counter()
for i in train[train['v1'] == 'spam']['v2']:
vocab_spam += Counter(i.split(' '))

vocab_ham = Counter()
for i in train[train['v1'] == 'ham']['v2']:
vocab_ham += Counter(i.split(' '))

In [17]:

test_sentence = 'to a a spam msg'

localhost:8888/notebooks/Exp-9.ipynb 1/2
5/23/22, 8:08 PM Exp-9 - Jupyter Notebook

In [18]:

intial_spam_guess = train[train['v1'] == 'spam'].shape[0]/train.shape[0]


intial_ham_guess = train[train['v1'] == 'ham'].shape[0]/train.shape[0]

In [22]:

d = {'TP': 0, 'FP': 0, 'TN': 0, 'FN': 0}

for (sentence, label) in zip(X_test, y_test):


spam_score = intial_spam_guess
ham_score = intial_ham_guess
for word in sentence.split(' '):
spam_score *= vocab_spam.get(word, 1)/sum(vocab_spam.values())
ham_score *= vocab_ham.get(word, 1)/sum(vocab_ham.values())

if spam_score > ham_score:


pred = 'spam'
else:
pred = 'ham'

if label == pred:
if label == 'spam':
d['TP'] += 1
else:
d['TN'] += 1
else:
if label == 'spam':
d['FP'] += 1
else:
d['FN'] += 1

In [29]:

print(f"Accuracy: {(d['TP'] + d['TN'])/sum(d.values())}")


print(f"Precision: {d['TP']/(d['TP'] + d['FP'])}")
print(f"Recall: {d['TP']/(d['TP'] + d['FN'])}")

Accuracy: 0.8810432569974554

Precision: 0.9808612440191388

Recall: 0.5283505154639175

In [ ]:

localhost:8888/notebooks/Exp-9.ipynb 2/2

You might also like