Python 层次聚类+K-means算法+可视化 数据挖掘

层次聚类:

import pandas as pd
import scipy.cluster.hierarchy as sch
from sklearn.cluster import AgglomerativeClustering
from sklearn.preprocessing import MinMaxScaler
from matplotlib import pyplot as plt
# 读取数据
data = pd.read_csv(r'C:/Users/24224/Desktop/countries/test2.csv',index_col=0)
# print(data.head())
df = MinMaxScaler().fit_transform(data)
# 建立模型
model = AgglomerativeClustering(n_clusters=4)
model.fit(df)
data['类别标签'] = model.labels_
print(data.head())
# 画图
ss = sch.linkage(df,method='ward')
sch.dendrogram(ss)
plt.show()

预处理,填充缺失值

# -*- coding: utf-8 -*-
#导入相应的包
import scipy
import scipy.cluster.hierarchy as sch
from scipy.cluster.vq import vq,kmeans,whiten
import numpy as np
import math
import matplotlib.pylab as plt
import pandas as pd
#导入pandas包
df= pd.read_csv("C:/Users/24224/Desktop/countries/countries of the world.csv")
df=df.drop(['Country','Region'],axis=1)
df = df.apply(pd.to_numeric, errors='coerce')
df = df.fillna(df.mean())
df.isnull()
"""
df['Net migration'].fillna(df['Net migration'].mean())
df['Infant mortality (per 1000 births)'].fillna(df['Infant mortality (per 1000 births)'].mean())
df['GDP ($ per capita)'].fillna(df['GDP ($ per capita)'].mean())
df['Literacy (%)'].fillna(df['Literacy (%)'].mean())
df['Phones (per 1000)'].fillna(df['Phones (per 1000)'].mean())
df['Arable (%)'].fillna(df['Arable (%)'].mean())
df['Crops (%)'].fillna(df['Crops (%)'].mean())
df['Other (%)'].fillna(df['Other (%)'].mean())
df['Climate'].fillna(df['Climate'].mean())
df['Birthrate'].fillna(df['Birthrate'].mean())
df['Deathrate'].fillna(df['Deathrate'].mean())
df['Agriculture'].fillna(df['Agriculture'].mean())
df['Industry'].fillna(df['Industry'].mean())
df['Service'].fillna(df['Service'].mean())"""
df.to_csv("C:/Users/24224/Desktop/countries/test2.csv",index=False,sep=',')

k-means聚类:

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
# 加载数据
def loadDataSet(fileName):
    data = pd.read_csv(fileName,usecols=[17,6])
    return data
# 欧氏距离计算
def distEclud(x, y):
    return np.sqrt(np.sum((x - y) ** 2))  # 计算欧氏距离
# 为给定数据集构建一个包含K个随机质心的集合
def randCent(dataSet, k):
    m, n = dataSet.shape
    centroids = np.zeros((k, n))
    for i in range(k):
        index = int(np.random.uniform(0, m))  #
        centroids[i, :] = dataSet[index, :]
    return centroids
# k均值聚类
def KMeans(dataSet, k):
    m = np.shape(dataSet)[0]  # 行的数目
    # 第一列存样本属于哪一簇,第二列存样本的到簇的中心点的误差
    clusterAssment = np.mat(np.zeros((m, 2))) #生成矩阵
    clusterChange = True
    # 第1步 初始化centroids
    centroids = randCent(dataSet, k)
    while clusterChange: # while True
        clusterChange = False
        # 遍历所有的样本（行数）
        for i in range(m):
            minDist = 100000.0
            minIndex = -1# 遍历所有的质心
            # 第2步 找出最近的质心
            for j in range(k):
                # 计算该样本到质心的欧式距离
                distance = distEclud(centroids[j, :], dataSet[i, :]) # minDist = 100000.0  minIndex = -1
                if distance < minDist:
                    minDist = distance
                    minIndex = j #质心为j
# clusterAssment,第一列存样本属于哪一簇,第二列存样本的到簇的中心点的误差,
            # 第 3 步：更新每一行样本所属的簇
            if clusterAssment[i, 0] != minIndex: #clusterAssment = np.mat(np.zeros((m, 2))) #生成矩阵
                clusterChange = True
                clusterAssment[i, :] = minIndex, minDist ** 2
        # 第 4 步：更新k个质心
        for j in range(k):
            pointsInCluster = dataSet[np.nonzero(clusterAssment[:, 0].A == j)[0]]  # 获取簇类所有的点,非零元素
            centroids[j, :] = np.mean(pointsInCluster, axis=0)  # 对矩阵的行求均值
    print("Congratulations,cluster complete!")
    return centroids, clusterAssment
def showCluster(dataSet, k, centroids, clusterAssment):
    m, n = dataSet.shape
    if n != 2:
        print("数据不是二维的")
        return 1
    mark = ['or', 'ob', 'og', 'ok', '^r', '+r', 'sr', 'dr', ', 'pr']
    if k > len(mark):
        print("k值太大了,不够标记了")
        return 1
    # 绘制所有的样本
    for i in range(m):
        markIndex = int(clusterAssment[i, 0])
        plt.plot(dataSet[i, 0], dataSet[i, 1], mark[markIndex])
    mark = ['Dr', 'Db', 'Dg', 'Dk', '^b', '+b', 'sb', 'db', ', 'pb']
    # 绘制质心
    for i in range(k):
        plt.plot(centroids[i, 0], centroids[i, 1], mark[i])
    plt.show()
dataSet = loadDataSet("C:/Users/24224/Desktop/countries/test2.csv")
k = 4
centroids, clusterAssment = KMeans(dataSet.values, k)
showCluster(dataSet.values, k, centroids, clusterAssment)

可视化: