SMOTE

 import pandas as pd

import numpy as np

df=pd.read_csv("/content/sample_data/Mall_Customers.csv")

df.head()

print(df.shape)

print(df['Gender'].value_counts())

from sklearn.preprocessing import LabelEncoder

# Convert Gender to numerical values

le = LabelEncoder()

df['Gender'] = le.fit_transform(df['Gender'])

from imblearn.over_sampling import SMOTE

import pandas as pd

X = df.drop(columns=['Gender'])

y = df['Gender']  # Target: Gender column

# Apply SMOTE

smote = SMOTE(random_state=42)

X_resampled, y_resampled = smote.fit_resample(X, y)

df_resampled = pd.DataFrame(X_resampled, columns=X.columns)

df_resampled['Gender'] = y_resampled  # Add balanced Gender column back

from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()

df_resampled[['Annual Income (k$)', 'Age', 'Spending Score (1-100)']] = scaler.fit_transform(

    df_resampled[['Annual Income (k$)', 'Age', 'Spending Score (1-100)']]

)

from sklearn.cluster import KMeans

# Apply K-Means

kmeans = KMeans(n_clusters=4, random_state=42)

df_resampled['Cluster'] = kmeans.fit_predict(df_resampled[['Annual Income (k$)', 'Age', 'Spending Score (1-100)']])

import matplotlib.pyplot as plt

plt.scatter(df_resampled['Annual Income (k$)'], df_resampled['Spending Score (1-100)'], c=df_resampled['Cluster'], cmap='viridis')

plt.xlabel('Annual Income (k$)')

plt.ylabel('Spending Score (1-100)')

plt.title('Customer Clusters (After SMOTE & K-Means)')

plt.colorbar(label='Cluster')

plt.show()