[ML] K-means clustering 예제

dataset

CustomerDataSet.xlsx

0.01MB

import pandas as pd

data = pd.read_excel('/content/CustomerDataSet.xlsx')
data

import pandas as pd
from sklearn.cluster import KMeans
import matplotlib.pyplot as plt
from sklearn.preprocessing import MinMaxScaler
%matplotlib inline

data = pd.read_excel('/content/CustomerDataSet.xlsx')
data

processed_data = data.copy()

#데이터 전처리 - 정규화를 위한 작업
scaler = MinMaxScaler()

scaler.fit(processed_data[['ItemsBought','ItemsReturned']])
processed_data[['ItemsBought','ItemsReturned']] = scaler.transform(processed_data[['ItemsBought','ItemsReturned']])

#화면 figure 생성
plt.figure(figsize=(10,6))

for i in range(1,7):
  model = KMeans(n_clusters=i)
  model.fit(processed_data[['ItemsBought','ItemsReturned']])
  model_predict = model.predict(processed_data[['ItemsBought','ItemsReturned']])
  centroids = model.cluster_centers_
  print(model.cluster_centers_)
plt.figure(figsize=(16,8))
plt.xlabel('ItemsBought')
plt.ylabel('ItemsReturned')
#클러스터링 그리기
plt.scatter(processed_data['ItemsBought'],processed_data['ItemsReturned'],c=model_predict)
plt.scatter(centroids[:,0],centroids[:,1],c='black',s=150,alpha=0.8)
plt.show()

#플로팅하기
plt.scatter(data['ItemsBought'],data['ItemsReturned'],c=model_predict)

#우편번호로 범례달기
for(index, c_id, bought, returned, zip_code, product) in data.itertuples():
  plt.annotate(zip_code,(bought+0.6,returned+0.6))

plt.xlabel('ItemsBought')
plt.ylabel('ItemsReturned')

plt.show()

%matplotlib inline

model=KMeans(n_clusters=3)
model.fit(processed_data[['ItemsBought','ItemsReturned']])
model_predict = model.predict(processed_data[['ItemsBought','ItemsReturned']])
centroids = model.cluster_centers_
print(model.cluster_centers_)
plt.figure(figsize=(16,8))
plt.xlabel('ItemsBought')
plt.ylabel('ItemsReturned')
#클러스터링 그리기
plt.scatter(processed_data['ItemsBought'],processed_data['ItemsReturned'],c=model_predict)
plt.scatter(centroids[:,0],centroids[:,1],c='black',s=150,alpha=0.8)
plt.show()

#클러스터 0로 분류된 데이터를 추출해보자
data[model_predict==0]

#클러스터 1로 분류된 데이터 추출
data[model_predict==1]

#클러스터 2로 분류된 데이터 추출
data[model_predict==2]

#create a scatter plot
plt.scatter(processed_data['ItemsBought'],processed_data['ItemsReturned'],c=model_predict)

#각 클러스터와 Product ID의 관계를 시각화
for index,c_id,bought,returned,zip_code,product in processed_data.itertuples():
  plt.annotate("Cluster{}:{}".format(model_predict[index],product),(bought,returned))

plt.xlabel('ItemsBought')
plt.ylabel('ItemsReturned')
plt.show()

#create a scatter plot
plt.scatter(data['ItemsBought'],data['ItemsReturned'],c=model_predict)

#각 클러스터와 지역과의 관계 시각화
for (index,c_id,bought,returned,zip_code,product) in data.itertuples():
  plt.annotate(zip_code,(bought+0.6,returned+0.6))

plt.xlabel('ItemsBought')
plt.ylabel('ItemsReturned')

plt.show()

https://github.com/erica00j/machinelearning/blob/main/Kmeans_Customer1.ipynb