Spam Classification using Multinomial Naive Bayes | Text classification using Multinomial Naive Bayes Example

Question

Build a spam filter using Python and the multinomial Naive Bayes algorithm. 

Check Spam or Ham? Email Classifier Using Python using MultinomialNB.

Dataset click here.

Answer 1

ASSIGNMENT 15

import numpy as np

import pandas as pd

import sklearn

data=pd.read_csv("/content/spam.csv")

data.head()

from sklearn.feature_extraction.text import CountVectorizer

CV=CountVectorizer()

xtrain=CV.fit_transform(data.Message)

xtest=CV.transform(data.Category)

xtrain.shape

from sklearn.feature_extraction.text import TfidfTransformer

tfidf=TfidfTransformer()

xtrain_tfidf=tfidf.fit_transform(xtrain)

xtest_tfidf=tfidf.transform(xtest)

xtrain_tfidf.shape

df_idf=pd.DataFrame(tfidf.idf_,index=CV.get_feature_names(),columns=["IDF_WEIGHTS"])

df_idf.sort_values(by=["IDF_WEIGHTS"]).head(10)

from sklearn.naive_bayes import MultinomialNB

model=MultinomialNB()

model.fit(xtrain_tfidf,data.Category)

ypred=model.predict(xtest_tfidf)

ypred

from sklearn.metrics import accuracy_score

accuracy_score(ypred,data.Category)*100