如何使用 matplotlib 绘制由 KMeans 生成的集群？

Question

我使用 KMeans 进行聚类，如下所示，但我不知道在散点图中绘制我的聚类。 或者也喜欢这个情节 我的代码是：

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 = 2
model = KMeans(n_clusters=true_k, init='k-means++', max_iter=100, n_init=1)
model.fit(X)

Answer 1

如果我正确理解了您的问题，我想您可能想要做这样的事情？在转换为聚类距离空间后，我绘制了数据，按标签着色。

import matplotlib as mpl
import matplotlib.pyplot as plt
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans
from sklearn.metrics import adjusted_rand_score
import pandas as pd

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."]

df = pd.DataFrame(documents) # read in your data with pd.read_csv or if in list form like above do this
vectorizer = TfidfVectorizer(stop_words='english')

X = vectorizer.fit_transform(df[0].values.astype('U')) # make sure you have unicode strings [0] is the column of the sentences


true_k = 2
model = KMeans(n_clusters=true_k, init='k-means++', max_iter=2000, n_init=20)
Xt = model.fit_transform(X)


  # things with tf-idf score 
X = X.toarray()
fns =  np.array(vectorizer.get_feature_names()) # feature names/ordered by index

# retrieve labels with max score
labels = model.labels_ 
d = []
for n in sorted(np.unique(labels)):
    t = X[(labels == n)].sum(axis=0) #max tf/idf score cumulative/cluster
    words = fns[t == t.max()]
    d.append(",".join(words))

t = Xt.T # cluster distance space X transpose to be plotted with mpl

### plot the clusters
fig, ax = plt.subplots(1,1)

cluster_color_dict = {0:'purple', 1 :'blue'} # change these to desired colors
for i in range(len(t[0])):

    ax.scatter(t[0][i], t[1][i], c= cluster_color_dict[labels[i]], edgecolors='grey', lw = 0.5, s = 200)

p1 = [] # legend patches 
for i in range(2):
    print i
    h = ax.scatter([],[], c= cluster_color_dict[i], 
                    edgecolors= 'grey', lw = 0.5, s = 80, label = d[i]) 
    p1.append(h)

l1 = ax.legend(handles = p1, title= 'cluster',  bbox_to_anchor = (1,1), loc = 'upper left')