算法

The practical result shows that this algorithm is easy to be realized in project by applying all sots of existing algorithms in computer system not depending of special fuzzy control module.

实践表明：该算法实现过程简捷，可直接在计算机控制系统中运用各种现有的算法实现，不依赖于专有的模糊控制算法模块，有利于工程应用。

And the algorithm has been proved by simulation test and practical process. The practical result shows that this algorithm is easy to be realized in project by applying all sots of existing algorithms in computer system not depending of special fuzzy control module.

实践表明：该算法实现过程简捷，可直接在计算机控制系统中运用各种现有的算法实现，不依赖于专有的模糊控制算法模块，有利于工程应用。

An inductive learning method for classification based on information theory is proposed, which is a maximum contribution classification method according to the theory of channel transmission in information theory, the definition contribution is developed based on probability distribution of classified space, probability transfer matrices of classified space and feature space and mutual information, then entities is classified by the maximum contribution method.

提出了一种基于信息论的归纳分类学习算法－最大贡献分类算法。该算法根据信息论中信道传输的理论，把类别空间和各特征空间的概率转移矩阵以及互信息结合起来，定义了贡献的概念，从而得出了对给定的观测值，用它的最大贡献作为对实体进行分类的规则。

Based on a review of literatures, this paper described methods for performing spectral decomposition.

在对目前常用的几种谱分解算法介绍的基础上，结合其中一些方法的谱分解结果，对各种算法的时间、频率分辨率进行了对比分析，指出了各类算法的局限性，并针对不同用途指出了适用的谱分解方法。

Experiments show that this method performances a faster and more stable convergence than dwMDS algorithm which is based on SMACOF algorithm.

实验表明该迭代算法收敛快速且稳定，比基于SMACOF算法的dwMDS算法在定位精度上有明显的提高。

The proposed algorithm has advantages to the global searching ability of genetic algorithm and the stable convergence of dimidiate searching root method.

结合传统遗传算法良好的全局搜索能力和二分法可靠的收敛性，本文提出了一种改进的遗传算法－层次压缩遗传算法。

The new algorithm has better performance in stable convergence rate, convergence speed, and threshold sensitivity compared with traditional fuzzy clustering algorithm.

该算法在收敛稳定性、收敛速度、初值敏感性方面，相对于传统模糊聚类算法有了明显改善，相关实验表明，该算法具备高效、稳定的工程应用价值。

The success of star map match is judged according to compute the EMD distance of two stars.

针对现有的识别算法存在识别速度慢，识别成功率不高，提出了一种新的星图识别算法，该算法根据恒星星等，利用锚定的方法，使待识别星图中的星落于导航星图中的星的ε邻域，对于未落于ε邻域的星，建立有向赋权二部图，进行完美匹配，根据计算星对之间EMD距离，来确定星图的匹配是否成功。

The new approach improves the step length and the convergence is analyzed. Numerical examples are given to illustrate the effectiveness of the proposed method.

在此基础上提出了求解只带有等式约束的凸二次规划问题的新算法，该算法改进了搜索步长，分析了收敛性，通过数值算例验证了算法的有效性和优越性。

Experiments and analysis are given about our algorithm, such as the analysis of selecting rules of parameters, step length, the complexity of the algorithm, and the comparison with other algorithms.

4针对我们提出的解决Job—Shop调度问题的神经网络方法做了实验分析，内容包括从实验角度分析算法中参数的选择方法与步长的选择方法；算法的复杂性分析以及与其它算法的比较。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。