资源列表
GMMPEM
- 代码给出了高斯分布下的EM算法的设计与实现-Code gives the design and implementation of a Gaussian distribution under the EM algorithm
Batch-Gradient-Descent
- 分别使用了批梯度下降法和牛顿法进行线性回归的测试。-Respectively the batch gradient descent and Newton s method of linear regression tests.
pocket-PLA
- 贪婪感知器算法。R语言实现版本，每次会把最优的分类抓在手上-pocket perceptron learning of algorithm
Naive-PLA
- 感知器学习算法。R语言实现版本。自己写的，给大家一个参考-perceptron learning of algorithm
Mymeans
- k-means（k均值聚类）,使用R语言实现，分类的准确度跟自带的差不多-k-means written by myself
myknn
- 自己写的knn算法，加入了剪辑近邻法,可以对类别交界处进行剪辑-knn written by myself
extrema
- 解决EMD分解后端点不正确的现象，该算法能有效解决这一问题-EMD decomposition procedure to solve the problem endpoint, can effectively solve the endpoint of decomposition phenomenon incorrect
LDM
- 对SVM分类方法进行的一种改进方法。将其中的margin改变。-SVM classification method for an improved method. The margin will be one change.
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.
shortest-path-of-N
- N最短路径是K最短路径（KSP）的变种，KSP指的是DAG中单源路径中前K条最短的路径- N shortest path is the shortest path (KSP) of K, KSP refers to the DAG in the single source path of the shortest path in the former K
Precision_Recall_F1-Measure
- 信息检索和自然语言处理中经常会使用这些参数:准确率（Precision）、召回率（Recall）以及综合评价指标（F1-Measure ） -These parameters are often used in information retri and natural language processing (Precision), recall rate (Recall), and comprehensive uation index (F1-Measure).
Bias_algorithm_java
- 贝叶斯算法java实现,在贝叶斯算法思想基础上做改进，提供文本分类效率-Bias algorithm java implementation, based on the idea of Bias algorithm to improve the efficiency of text classification