[딥러닝] DBScan

DBSCAN

 : 밀도 기반 클러스터링 : kmeans와 달리 k를 지정하지않음

 - 이론

untitledtblog.tistory.com/146

[데이터 마이닝] DBSCAN과 밀도 기반 클러스터링

 

  - K-Means / DBSCAN 동작 보기

https://www.naftaliharris.com/blog/visualizing-dbscan-clustering/

Visualizing DBSCAN Clustering

 

 * cluster6_dbscan

import matplotlib.pyplot as plt
import numpy as np
from sklearn.datasets import make_moons
from sklearn.cluster import KMeans, DBSCAN, AgglomerativeClustering

x, y = make_moons(n_samples = 200, noise = 0.05, random_state=0)
print(x)
print(y)
plt.scatter(x[:, 0], x[:, 1])
plt.show()

from sklearn.datasets import make_moons

make_moons(n_samples = 샘플수, noise = 노이즈 정도, random_state=난수seed) : 

def plotResult(x, y):
    plt.scatter(x[y == 0, 0], x[y == 0, 1], c='blue', marker='o', label='clu-1')
    plt.scatter(x[y == 1, 0], x[y == 1, 1], c='red', marker='s', label='clu-2')
    plt.legend()
    plt.show()

 

 - KMEANS 사용 : 완전한 분리 X

km = KMeans(n_clusters=2, random_state=0)
pred1 = km.fit_predict(x)

plotResult(x, pred1)

 - DBSCAN 사용

dm = DBSCAN(eps=0.2, min_samples=5, metric = 'euclidean')
pred2 = dm.fit_predict(x)
print(pred2)

plotResult(x, pred2)