Assignment no 2 _ML
October 8, 2024
[36]: #ASSIGNMENT NO 2
#Use K-Nearest Neighbors and Support Vector Machine for classification. Analyze␣
↪their performance.
#Dataset link: The emails.csv dataset on the Kaggle https://www.kaggle.com/
↪datasets/balaka18/email-spam-classification-dataset-csv
[37]: import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
[38]: df=pd.read_csv('/home/pc13/Documents/Email/emails.csv')
[51]: df.head() #it returns the first five rows of the DataFrame df
[51]: Email No. the to ect and for of a you hou … connevey jay \
0 Email 1 0 0 1 0 0 0 2 0 0 … 0 0
1 Email 2 8 13 24 6 6 2 102 1 27 … 0 0
2 Email 3 0 0 1 0 0 0 8 0 0 … 0 0
3 Email 4 0 5 22 0 5 1 51 2 10 … 0 0
4 Email 5 7 6 17 1 5 2 57 0 9 … 0 0
valued lay infrastructure military allowing ff dry Prediction
0 0 0 0 0 0 0 0 0
1 0 0 0 0 0 1 0 0
2 0 0 0 0 0 0 0 0
3 0 0 0 0 0 0 0 0
4 0 0 0 0 0 1 0 0
[5 rows x 3002 columns]
[52]: df.info() #df.info() function in pandas provides a concise summary of a␣
↪DataFrame
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 5172 entries, 0 to 5171
Columns: 3002 entries, Email No. to Prediction
1
dtypes: int64(3001), object(1)
memory usage: 118.5+ MB
[53]: df.isnull().sum() #The df.isnull().sum() function in pandas is used to check␣
↪for missing (null) values in a DataFrame.
[53]: 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
[54]: X = df.iloc[:, 1:-1].values
y = df.iloc[:, -1].values
#X and y are being created from a pandas DataFrame df using the iloc method,␣
↪which is used for integer-location based indexing
#X typically represents the feature set (input data) used for training a␣
↪machine learning model.
#y usually represents the target variable (output data) that the model aims to␣
↪predict.
[55]: from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30,␣
↪random_state=101)
#the train_test_split function from the sklearn.model_selection module is used␣
↪to split the dataset into training and testing sets
[56]: from sklearn.preprocessing import StandardScaler
sc_X = StandardScaler()
X_train = sc_X.fit_transform(X_train)
X_test = sc_X.transform(X_test)
#The StandardScaler from the sklearn.preprocessing module is used to␣
↪standardize the feature set
[57]: from sklearn.neighbors import KNeighborsClassifier
classifier = KNeighborsClassifier(n_neighbors=5)
classifier.fit(X_train, y_train)
#using the KNeighborsClassifier from the sklearn.neighbors module to create and␣
↪train a K-Nearest Neighbors (KNN) classifier.
2
[57]: KNeighborsClassifier()
[58]: #KNeighborsClassifier() is a class in the sklearn.neighbors module of the␣
↪scikit-learn library, which implements the K-Nearest Neighbors (KNN)␣
↪algorithm for classification tasks
[59]: y_pred = classifier.predict(X_test)
#In this line of code, y_pred = classifier.predict(X_test), you're using the␣
↪trained K-Nearest Neighbors classifier to make predictions on the test set
[60]: from sklearn.metrics import confusion_matrix, accuracy_score
cm = confusion_matrix(y_test, y_pred)
#using functions from sklearn.metrics to evaluate the performance of your␣
↪K-Nearest Neighbors classifier by generating a confusion matrix.
[61]: cm
#The variable cm contains the confusion matrix generated by the␣
↪confusion_matrix function.
[61]: array([[866, 248],
[ 16, 422]])
[49]: from sklearn.metrics import classification_report
cl_report=classification_report(y_test,y_pred)
print(cl_report)
#Generating a classification report using the classification_report function␣
↪from the sklearn.metrics module
precision recall f1-score support
0 0.98 0.78 0.87 1114
1 0.63 0.96 0.76 438
accuracy 0.83 1552
macro avg 0.81 0.87 0.81 1552
weighted avg 0.88 0.83 0.84 1552
[50]: print("Accuracy Score for KNN : ", accuracy_score(y_pred,y_test))
Accuracy Score for KNN : 0.8298969072164949
[62]: from sklearn.svm import SVC
from sklearn.metrics import accuracy_score
#importing the Support Vector Classifier (SVC) from the sklearn.svm module and␣
↪the accuracy_score function from sklearn.metrics
3
[69]: svc = SVC(C=1.0,kernel='rbf',gamma='auto')
svc.fit(X_train,y_train)
y_pred2 = svc.predict(X_test)
#you're creating and training a Support Vector Classifier (SVC) using the␣
↪Radial Basis Function (RBF) kernel
[64]: from sklearn.metrics import confusion_matrix, accuracy_score
#generating a confusion matrix for the predictions made by the Support Vector␣
↪Classifier (SVC
cm = confusion_matrix(y_test, y_pred2)
#Creating the Confusion Matrix
[70]: cm
#The variable cm contains the confusion matrix generated from your SVC model's␣
↪predictions
[70]: array([[1106, 8],
[ 95, 343]])
[67]: print("Accuracy Score for SVC : ", accuracy_score(y_pred2,y_test))
Accuracy Score for SVC : 0.9336340206185567
[71]: from sklearn.metrics import classification_report
cl_report=classification_report(y_test,y_pred2)
print(cl_report)
#generating a classification report for the predictions made by your Support␣
↪Vector Classifier (SVC)
precision recall f1-score support
0 0.92 0.99 0.96 1114
1 0.98 0.78 0.87 438
accuracy 0.93 1552
macro avg 0.95 0.89 0.91 1552
weighted avg 0.94 0.93 0.93 1552
[ ]:
