Email Spam Classification.ipynb

Kernel: Python 3.8.6 64-bit

import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score
from sklearn.neighbors import KNeighborsClassifier

df = pd.read_csv("./emails.csv")

df.head()

df.isnull().sum()

Email No.     0
the           0
to            0
ect           0
and           0
             ..
military      0
allowing      0
ff            0
dry           0
Prediction    0
Length: 3002, dtype: int64

X = df.iloc[:,1:3001]
X

Y = df.iloc[:,-1].values
Y

array([0, 0, 0, ..., 1, 1, 0], dtype=int64)

train_x,test_x,train_y,test_y = train_test_split(X,Y,test_size = 0.25)

svc = SVC(C=1.0,kernel='rbf',gamma='auto')         
# C here is the regularization parameter. Here, L2 penalty is used(default). It is the inverse of the strength of regularization.
# As C increases, model overfits.
# Kernel here is the radial basis function kernel.
# gamma (only used for rbf kernel) : As gamma increases, model overfits.
svc.fit(train_x,train_y)
y_pred2 = svc.predict(test_x)
print("Accuracy Score for SVC : ", accuracy_score(y_pred2,test_y))

Accuracy Score for SVC :  0.8979118329466357

X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size = 0.2, random_state=42)

knn = KNeighborsClassifier(n_neighbors=7)

knn.fit(X_train, y_train)

print(knn.predict(X_test))

[0 0 1 ... 0 1 0]

print(knn.score(X_test, y_test))

0.8685990338164251