聚類是一種機(jī)器學(xué)習(xí)技術(shù)，它涉及到數(shù)據(jù)點(diǎn)的分組。給定一組數(shù)據(jù)點(diǎn)，我們可以使用聚類算法將每個數(shù)據(jù)點(diǎn)劃分為一個特定的組。理論上，同一組中的數(shù)據(jù)點(diǎn)應(yīng)該具有相似的屬性和/或特征，而不同組中的數(shù)據(jù)點(diǎn)應(yīng)該具有高度不同的屬性和/或特征。聚類是一種無監(jiān)督學(xué)習(xí)的方法，是許多領(lǐng)域中常用的統(tǒng)計數(shù)據(jù)分析技術(shù)。

常用的算法包括K-MEANS、高斯混合模型（Gaussian Mixed Model，GMM）、自組織映射神經(jīng)網(wǎng)絡(luò)（Self-Organizing Map，SOM）

重點(diǎn)給大家介紹Python實(shí)現(xiàn)DBSCAN聚類算法并通過簡單樣例測試。

發(fā)現(xiàn)高密度的核心樣品并從中膨脹團(tuán)簇。

Python代碼如下：

# -*- coding: utf-8 -*-'''Demo of DBSCAN clustering algorithmFinds core samples of high density and expands clusters from them.'''print(__doc__)# 引入相關(guān)包import numpy as npfrom sklearn.cluster import DBSCANfrom sklearn import metricsfrom sklearn.datasets.samples_generator import make_blobsfrom sklearn.preprocessing import StandardScalerimport matplotlib.pyplot as plt# 初始化樣本數(shù)據(jù)centers = [[1, 1], [-1, -1], [1, -1]]X, labels_true = make_blobs(n_samples=750, centers=centers, cluster_std=0.4, random_state=0)X = StandardScaler().fit_transform(X)# 計算DBSCANdb = DBSCAN(eps=0.3, min_samples=10).fit(X)core_samples_mask = np.zeros_like(db.labels_, dtype=bool)core_samples_mask[db.core_sample_indices_] = Truelabels = db.labels_# 聚類的結(jié)果n_clusters_ = len(set(labels)) - (1 if -1 in labels else 0)n_noise_ = list(labels).count(-1)print(’Estimated number of clusters: %d’ % n_clusters_)print(’Estimated number of noise points: %d’ % n_noise_)print('Homogeneity: %0.3f' % metrics.homogeneity_score(labels_true, labels))print('Completeness: %0.3f' % metrics.completeness_score(labels_true, labels))print('V-measure: %0.3f' % metrics.v_measure_score(labels_true, labels))print('Adjusted Rand Index: %0.3f' % metrics.adjusted_rand_score(labels_true, labels))print('Adjusted Mutual Information: %0.3f' % metrics.adjusted_mutual_info_score(labels_true, labels, average_method=’arithmetic’))print('Silhouette Coefficient: %0.3f' % metrics.silhouette_score(X, labels))# 繪出結(jié)果unique_labels = set(labels)colors = [plt.cm.Spectral(each) for each in np.linspace(0, 1, len(unique_labels))]for k, col in zip(unique_labels, colors): if k == -1:col = [0, 0, 0, 1] class_member_mask = (labels == k) xy = X[class_member_mask & core_samples_mask] plt.plot(xy[:, 0], xy[:, 1], ’o’, markerfacecolor=tuple(col), markeredgecolor=’k’, markersize=14) xy = X[class_member_mask & ~core_samples_mask] plt.plot(xy[:, 0], xy[:, 1], ’o’, markerfacecolor=tuple(col), markeredgecolor=’k’, markersize=6)plt.title(’Estimated number of clusters: %d’ % n_clusters_)plt.show()

測試結(jié)果如下：

最終結(jié)果繪圖：

具體數(shù)據(jù)：

以上就是Python實(shí)現(xiàn)DBSCAN聚類算法（簡單樣例測試）的詳細(xì)內(nèi)容，更多關(guān)于Python聚類算法的資料請關(guān)注好吧啦網(wǎng)其它相關(guān)文章！