证据积累聚类集成算法（Evidence Accumulation Clustering）代码复现与实验

1. 基本环境

运行环境：
- Python 3.7 +
- Jupyter NoteBook
- 处理器：2.6 GHz 六核Intel Core i7

2. 聚类集成代码

# 导入包
import pandas as pd
import numpy as np
import math
import random
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans, AgglomerativeClustering
from sklearn import datasets
from scipy.cluster.hierarchy import dendrogram, linkage, fcluster
from sklearn import metrics
import scipy.spatial.distance as dist

# Kmeans-聚类算法
def kmeans_cluster(data, **kwargs):  # K-means
    if kwargs is None:
        kwargs = {
   'k': 8}
    kmean = KMeans(n_clusters=kwargs['k'])
    return kmean.fit_predict(data)

# 更新邻近度矩阵
# similar_mat: 原邻近度矩阵
# cluster_ind: 类群的样本序号列表
def update_similar_mat(similar_mat, cluster_ind):
  if len(cluster_ind) <= 1:
    return similar_mat
  for ind, no_i in enumerate(cluster_ind): # ind 是下标， no_i 是值
    if no_i == cluster_ind[-1]:
      return similar_mat
    for no_j in cluster_ind[ind+1:]:
      similar_mat[no_i, no_j] += 1
      similar_mat[no_j, no_i] += 1

# 单聚类算法
def single_fit(para_list, data):
  k = random.randint(0, len(para_list)-1)
  cluster_result = kmeans_cluster(data, k=para_list[k])
  return cluster_result

# number 聚类次数
# X 原始数据
# result_dict 聚类结果
def ensemble_SL(number, X, result_dict):
  threshold = 0 # 阈值
  similar_mat = np.zeros((X.shape[0], X.shape[0]))
  for single_result in result_dict:
    k_num = list