Customer Segmentation

%matplotlib inline import pandas as pd import numpy as np import matplotlib.pyplot as plt print('Libraries imported')

Libraries imported

df_customers = pd.read_csv('../data/Mall_Customers.csv') df_customers.head()

CustomerIDint64

Genderobject

Male

Female

df_customers = df_customers.drop(['CustomerID'], axis=1) df_customers.head()

Genderobject

Ageint64

Male

Female

df_customers.isnull().sum()

from sklearn.preprocessing import StandardScaler

numerical_columns = ['Age', 'Annual Income (k$)', 'Spending Score (1-100)'] numerical_data = df_customers[numerical_columns] scaler = StandardScaler().fit(numerical_data.values) numerical_data = scaler.transform(numerical_data.values) scaled_num_data = pd.DataFrame(numerical_data, columns=numerical_columns) scaled_num_data.head()

Agefloat64

Annual Income (k$)float64

-1.4245687900521393

-1.7389991930659485

-1.2810354107017978

-1.7389991930659485

-1.3528021003769686

-1.7008297638941763

-1.1375020313514563

-1.7008297638941763

-0.5633685139500905

-1.6626603347224038

gender = df_customers['Gender'] df = scaled_num_data.join(gender) df = pd.get_dummies(df) df = df.drop(['Gender_Male'], axis=1) df.head()

Agefloat64

Annual Income (k$)float64

-1.4245687900521393

-1.7389991930659485

-1.2810354107017978

-1.7389991930659485

-1.3528021003769686

-1.7008297638941763

-1.1375020313514563

-1.7008297638941763

-0.5633685139500905

-1.6626603347224038

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

Agefloat64

Annual Income (k$)float64

count

200

mean

-1.021405182655144e-16

-2.1316282072803005e-16

std

1.002509414234171

min

-1.49633547972731

-1.7389991930659485

25%

-0.7248435657192246

-0.727509320013981

50%

-0.20453506557423673

0.035879263421465926

75%

0.728431900202983

0.6656748447557098

max

2.2355323833815683

2.9176711658902788

covariance_matrix = np.cov(df[['Age', 'Annual Income (k$)', 'Spending Score (1-100)']].T) covariance_matrix

import seaborn as sns plt.figure(figsize=(10,10)) sns.set(font_scale=1.5) hm = sns.heatmap(covariance_matrix, cbar=True, annot=True, square=True, fmt='.2f', annot_kws={'size': 12}, yticklabels=['Age', 'Annual Income (k$)', 'Spending Score (1-100)'], xticklabels=['Age', 'Annual Income (k$)', 'Spending Score (1-100)'])

from sklearn.decomposition import PCA pca = PCA(n_components=4) principal_components = pca.fit_transform(df) features = range(pca.n_components_) plt.bar(features, pca.explained_variance_ratio_) plt.xlabel('PCA features') plt.ylabel('variance %') plt.xticks(features) print(pca.explained_variance_ratio_) PCA_components = pd.DataFrame(principal_components)

[0.4095285  0.3081674  0.20723465 0.07506945]

from sklearn.cluster import KMeans ks = range(1, 10) inertias = [] for k in ks: model = KMeans(n_clusters=k) model.fit(PCA_components.iloc[:,:3]) inertias.append(model.inertia_) plt.plot(ks, inertias, '-o', color='black') plt.xlabel('number of clusters, k') plt.ylabel('inertia') plt.xticks(ks) plt.show()

df = pd.read_csv('../data/Mall_Customers.csv') df = df.drop(['CustomerID'],axis=1) model = KMeans(n_clusters=4) model.fit(PCA_components.iloc[:,:3]) k_predict = model.predict(PCA_components.iloc[:,:3]) frame = pd.DataFrame(df) frame['cluster'] = k_predict frame.head()

Genderobject

Ageint64

Male

Female

avg_df = df.groupby(['cluster'], as_index=False).mean() avg_df

clusterint64

Agefloat64

25.43859649122807

53.98461538461538

32.875

39.36842105263158

fig, ax = plt.subplots(1,3,figsize=(20,5)) sns.barplot(x='cluster',y='Age',data=avg_df, ax=ax[0]) sns.barplot(x='cluster',y='Spending Score (1-100)',data=avg_df, ax=ax[1]) sns.barplot(x='cluster',y='Annual Income (k$)',data=avg_df, ax=ax[2])

df2 = pd.DataFrame(df.groupby(['cluster','Gender'])['Gender'].count()) df2

Genderint64

(0, 'Female')

(0, 'Male')

(1, 'Female')

(1, 'Male')

(2, 'Female')

(2, 'Male')

(3, 'Female')

(3, 'Male')

df3 = pd.DataFrame(df.groupby(['Gender'])['Gender'].count()) df3

Genderint64

Female

112

Male