算法信息论

该算法利用信息论的熵函数，根据MDL原理构造评价函数，在全部属性集合中寻找最优属性子集。

This algorithm uses the concept of entropy and MDLP to define the evaluation function of a feature subset and searches the best feature subset among all features using this function .

该算法从信息论的角度出发，综合利用图象信息熵及不变矩的特性来描述商标图象的形状特征。

Based on the theory of information , the method uses images ' entropy and invariant moments to capture the shape and spatial information of images .

阐述了阵列信号处理中广泛采用的用于来波方向（DOA）估计的多信号分类（MUSIC）算法原理及基于信息论准则的信源数目判别方法；

The paper introduces briefly first the principle of MUSIC algorithm widely used today to estimate the direction of arrivals ( DOA ) in array signal processing , and the method of estimating the number of signals basing on information theory ;

为此，采用市场值函数算法，它以信息论为基础，通过构造一个线性市场值函数来对客户进行排序，从而发现最具有潜在市场价值的客户群。

Therefore , based on the informational theory , the market value algorithm , is adopted to propose a linear market value function so as to discover the customers with the most potential market value .

同时也提出了MIBARK算法，该算法基于信息论的方法，用信息熵来定义属性的重要性。

In the process of attribute reduction , attribute reduction algorithms based on discernibility matrix is proposed .