python - Kmeans 返回的集群可视化

Question

我使用 KMeans 进行聚类，如下所示，但我不知道要像下图所示那样可视化聚类以查看客户的满意度。

代码:

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans
from sklearn.metrics import adjusted_rand_score

documents = ["This little kitty came to play when I was eating at a          restaurant.",
         "Merley has the best squooshy kitten belly.",
         "Google Translate app is incredible.",
         "If you open 100 tab in google you get a smileyface.",
         "Best cat photo I've ever taken.",
         "Climbing ninja cat.",
         "Impressed with google map feedback.",
         "Key promoter extension for Google Chrome."]

  vectorizer = TfidfVectorizer(stop_words='english')
  X = vectorizer.fit_transform(documents)

 true_k = 3
 model = KMeans(n_clusters=true_k, init='k-means++',  max_iter=100,n_init=1)
 model.fit(X)

Answer 1

假设您有办法知道 k-means 的哪个分区代表哪种情绪，您可以绘制饼图如下:

print(model.labels_)  # For illustration, you can see which sentence is in which cluster
# Here we get the proportions
nb_samples = [sum(model.labels_ == j) for j in range(true_k)]

# On the next line the order is RANDOM. I do NOT know which cluster represents what.
# The first label should represent samples in cluster 0, and so on
labels = 'positive', 'neutral', 'negative'
colors = ['gold', 'red', 'lightblue']  # Same size as labels

# Pie chart
plt.pie(nb_samples, labels=labels, colors=colors, autopct='%1.1f%%')
plt.axis('equal')
plt.show()

也不是，多次运行会根据哪个聚类代表哪个类别给出不同的结果。

这可以通过设置 numpy 随机种子来避免。

import numpy as np
np.random.seed(42)  # Or any other integer