naive bayes

multinomial.py

import nltk
nltk.download('stopwords')
import pandas as pd
import numpy as np
from nltk.corpus import stopwords
import string
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.preprocessing import LabelBinarizer
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import accuracy_score

df = pd.read_csv("spam.csv",encoding='iso8859_14')
df.drop(labels=df.columns[2:],axis=1,inplace=True)
df.columns=['target','text']

def clean_util(text):
    punc_rmv = [char for char in text if char not in string.punctuation]
    punc_rmv = "".join(punc_rmv)
    stopword_rmv = [w.strip().lower() for w in punc_rmv.split() if w.strip().lower() not in stopwords.words('english')]
    
    return " ".join(stopword_rmv)
df['text'] = df['text'].apply(clean_util)

cv = CountVectorizer()
X = cv.fit_transform(df['text']).toarray()
lb = LabelBinarizer()
y = lb.fit_transform(df['target']).ravel()

# Train Test Split
X_train, X_test, y_train, y_test = train_test_split(X,y)

clf = MultinomialNB(alpha=1)
clf.fit(X,y)
y_pred = clf.predict(X_test)