The Wahyudiharto's Blog

Data:

Questionnaire data from mall visitors contains sex, age, salary & shopping score (200 rows).

Mission:

How to predict the cluster group from given age & salary

Library used:

• Pandas
• Numpy
• Seaborn
  
• Matplotlib
• Scikit

Code:

import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans

url = 'https://raw.githubusercontent.com/kokocamp/vlog101/master/vlog101.csv'
vlog128 = pd.read_csv(url)
vlog128.info()

X = vlog128[['Usia','Gaji (juta)']]
y = vlog128['Skor Belanja (1-100)']

sns.scatterplot(x="Usia",y="Gaji (juta)",data=vlog128,color="red",alpha=0.5)

X = np.array(X)
n_clusters = 5
kmeans = KMeans(n_clusters)
kmeans.fit(X)
print(kmeans.cluster_centers_)

nc = []
for i in range(n_clusters):
  nc.append(i)

print(kmeans.labels_)
vlog128["kluster"] = kmeans.labels_
vlog128.head()

fig, ax = plt.subplots()
sct = ax.scatter(X[:,0],X[:,1], c = vlog128.kluster, marker = "o", alpha = 0.5)
centers = kmeans.cluster_centers_
ax.scatter(centers[:,0], centers[:,1], c=nc, s=200, alpha=0.5)
plt.title("Hasil Klustering K-Means")
plt.xlabel("Usia")
plt.ylabel("Gaji (juta)")
plt.show()

usia = input("Usia (thn): ")
usia = int(usia)
gaji = input("Gaji (juta): ")
gaji = int(gaji)
data = [usia,gaji]
hasil = kmeans.predict([data])
print("Prediksi Kluster (0-4): ", hasil)

fig, ax = plt.subplots()
sct = ax.scatter(X[:,0],X[:,1], c = vlog128.kluster, marker = "o", alpha = 0.5)
centers = kmeans.cluster_centers_
ax.scatter(centers[:,0], centers[:,1], c=nc, s=200, alpha=0.5)
plt.title("Hasil Klustering K-Means")
plt.xlabel("Usia")
plt.ylabel("Gaji (juta)")
plt.scatter(usia,gaji, c = "red", s=100)
plt.show()

I wrapped the scenario in a Youtube video below.

Labels: Programming, Python

PS: If you've benefit from this blog, you can support it by making a small contribution.