จัดกลุ่มข้อมูลด้วย K-Means โดยใช้ Python

import math
import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd
from sklearn.cluster import KMeans 
from kneed import KneeLocator
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import silhouette_score

import os
for dirname, _, filenames in os.walk('/kaggle/input'):
    for filename in filenames:
        print(os.path.join(dirname, filename))

wine_df = pd.read_csv("/kaggle/input/wine-dataset-for-clustering/wine-clustering.csv", )
wine_df

sc = StandardScaler()
wine_df_sc = sc.fit_transform(wine_df)

WSS = []

for i in range (1,15):
    kmeans = KMeans(n_clusters=i, n_init=10)
    kmeans.fit(wine_df_sc)
    WSS.append(kmeans.inertia_)

kl = KneeLocator(range(1, 15), WSS, curve="convex", direction="decreasing")
optimal_k = kl.elbow
print("Optimal number of clusters: ", optimal_k)
plt.figure(figsize=(10,6))
plt.title("Elbow Analysis", fontsize=18)
sns.lineplot(x=list(range(1,15)), y=WSS, color="blue", marker="o", markersize=16)
plt.xlabel("Number of clusters")
plt.ylabel("WSS")
plt.vlines(optimal_k, plt.ylim()[0], plt.ylim()[1], linestyles='dashed')
plt.show()

silhouette = []

for i in range(2,15):
    kmeans = KMeans(n_clusters=i, n_init=10)
    cluster_labels = kmeans.fit_predict(wine_df_sc)
    silhouette.append(silhouette_score(wine_df_sc,cluster_labels))   

plt.figure(figsize=(10,6))
plt.title("Silhouette Analysis", fontsize=18)
sns.lineplot(x=list(range(2,15)), y=silhouette, color="green", marker="o", markersize=16)
plt.xlabel("Number of clusters")
plt.ylabel("Silhouette Score")
plt.show()

print("Best Silhouette Clustering is : ", max(silhouette), "with ", silhouette.index(max(silhouette))+2)

kmeans = KMeans(n_clusters=3, n_init=10, random_state=51)
kmeans.fit(wine_df_sc)
wine_df['Cluster']=kmeans.labels_

wine_df.groupby('Cluster').describe(percentiles=[])

from sklearn.decomposition import PCA

data=wine_df.drop("Cluster",axis=1).values
data_sc=sc.fit_transform(data)
pca=PCA(n_components=2)
principal_components = pca.fit_transform(data_sc)
df_pca = pd.DataFrame(principal_components, columns=['PC1', 'PC2'])
df_pca['Cluster']=wine_df['Cluster']

plt.figure(figsize=(10, 7))  
plt.title('Clustering Results using PCA', fontsize=18)
sns.scatterplot(data=df_pca, x='PC1', y='PC2', hue="Cluster", palette="magma", s=100)
plt.grid()
plt.show()

cluster_means=wine_df.groupby('Cluster').mean()
cluster_means