资源列表
k_clique
- [X,Y,Z] = k_clique(k,A) Inputs: k - clique size A - adjacency matrix Outputs: X - detected communities Y - all cliques (i.e. complete subgraphs that are not parts of larger complete subgraphs) Z - k-clique matrix-k-clique alg
OPTICS-algorithm
- optics 算法作为基于密度的聚类算法的一种重要的改进,十分有借鉴的意义。-Called algorithm for clustering algorithm based on density is an important improvement, very have reference significance.
DeepLearning-master
- 深度学习的概念源于人工神经网络的研究。含多隐层的多层感知器就是一种深度学习结构。深度学习通过组合低层特征形成更加抽象的高层表示属性类别或特征,以发现数据的分布式特征表示。[1] 深度学习的概念由Hinton等人于2006年提出。基于深信度网(DBN)提出非监督贪心逐层训练算法,为解决深层结构相关的优化难题带来希望,随后提出多层自动编码器深层结构。此外Lecun等人提出的卷积神经网络是第一个真正多层结构学习算法,它利用空间相对关系减少参数数目以提高训练性能。[1] 深度学习是机器
mani
- 此代码是关于流形学习,数据降维,代码中含有的主要方法是PCA,KPCA,MDS,KMDS,Laplacian等等,且代码作了可视化处理,界面效果完美-This code is on the manifold learning, data dimensionality reduction, the main method code is contained in PCA, KPCA, MDS, KMDS, Laplacian, etc., and the code visualization ma
Data
- 《R统计与数据挖掘》一书中所有源码,可以直接编译- R statistics and data mining, a book all the source code, can be directly compiled
scaleForSVM
- 用SVM进行分类时,需要对原数据进行归一化处理。-When using SVM classification, the need for raw data normalized.
dataset
- 自己写的python canny算法,可以对图片进行边缘检测,要求见PPT,可根据需要调整代码-Write your own python canny algorithm that can detect the edge of the picture, asked to see the PPT, the code can be adjusted as needed
FP-Growth
- fp-growth算法的c++实现,比较简单,但不能生成关联规则。-fp-growth algorithm c++ implementation, relatively simple, but can not generate association rules.
ID3
- 决策树算法的实现,id3算法,内含实验数据以及报告-an example of id3 algorithm
SPPP3.0
- 包含了信号处理的各种算法程序,如传统的傅里叶变换等时域和时频域分析。-Contains a variety of signal processing algorithms
LSSVM
- 最小二乘支持向量机,程序粘到command window里,设定 2 两个参数,可以更改,以达到最优化-igam=0.001 isig2=0.001 [gam,sig2]=tunelssvm({X,Y, f ,igam,isig2, RBF_kernel },... [0.001 0.001 10000 10000], gridsearch ,{}, leaveoneout_lssvm ) type= function approximation kernel= RBF_
FCBF
- This a one of the data mining algorithm completly made by me. This is based on feature subset selection. One of the popular algorithm of feature selection named Fast Correlation based feature selection algorithm(FCBF).It s complete code is here with