聚类算法
Partitional Clustering Algorithm
k均值聚类
PAM(Partitioning Around Medoids)
CLARA(Clustering Large Applications)
CLARANS(Clustering Large Applications based on Randomlized Search)
Hierarchical Clustering Algorithm
BIRCH(balanced iterative reducing and clustering using hierachies)
CURE(Clustering Using Representatives)
ROCKS(A Robust Clustering Algorithm for Categorial Attributes)
Grid-based Clustering Algorithm
STING
WaveCluster
Model-based Clustering Algorithm
Gaussian model
Latent Dirichlet Allocation
Density-based Clustering Algorithm
DBSCAN聚类
- 参考博客
- 关键概念：核心对象，密度直达，密度可达，密度相连
- 缺点：需要自己确定eps，Minpts
AE-DBSCAN(可以自己决定radius，Eps,parameters)
- k-dist list:对于给定数据集D，k-dist list是数据集中对于每个点来说距离最近的k个点
- slope of a point with respect to another point:对于一个给定的$k-list=(k_1,k_2,…,k_n)$,$k_i$到$k_{i+1}$的斜率就是$k_ik_{i+1}$的斜率
OPTICS
- 参考博客
- 关键概念：核心距离，可达距离
DENCLUE
CLIQUE
关联规则
APriori算法
FP_Tree算法
决策树
ID3+C4.5
CART算法
多目标进化算法
- NSGA，NSGA-II，SPEA，SPEA-II，MOEA/D,DMOEA-$\epsilon C$
差分进化算法
粒子群算法
分布估计算法
local search
- tabu search
- 爬山法
