一种基于加权KNN的大数据集下离群检测算法

传统KNN算法是在基于距离的离群检测算法的基础上提出的一种在大数据集下进行离群点挖掘的算法,然而KNN算法只以最近的第k个邻居的距离作为判断是否是离群点的标准有时也失准确性.给出了一种在大数据集下基于KNN的离群点检测算法,即在传统KNN方法的基础上为每个数据点增加了权重,权重值为与最近的k个邻居的平均距离,离群点为那些与第k个邻居的距离最大且相同条件下权重最大的点.算法能提高离群点检测的准确性,通过实验验证了算法的可行性,并与传统KNN算法的性能进行了对比.     

An efficient accelerator for attribute reduction from incomplete data in rough set framework

Feature selection (attribute reduction) from large-scale incomplete data is a challenging problem in areas such as pattern recognition, machine learning and data mining. In rough set theory, feature selection from incomplete data aims to retain the discriminatory power of original features. To address this issue, many feature selection algorithms have been proposed, however, these algorithms are often computationally time-consuming. To overcome this shortcoming, we introduce in this paper a theoretic framework based on rough set theory, which is called positive approximation and can be used to accelerate a heuristic process for feature selection from incomplete data. As an application of the proposed accelerator, a general feature selection algorithm is designed. By integrating the accelerator into a heuristic algorithm, we obtain several modified representative heuristic feature selection algorithms in rough set theory. Experiments show that these modified algorithms outperform their original counterparts. It is worth noting that the performance of the modified algorithms becomes more visible when dealing with larger data sets.     

一种多智能体系统中的协作机制

在多智能体系统MAS中,自治的agent在完成任务时常常会出现任务重叠的“正交互”现象,从而使得系统效率低下。为了解决这一问题,文章提出了一种agent协作机制,采用最短路径的方法来找出agent最优的合作对象,优化整个系统处理问题的能力。实验证明该方法是有效的。     

Eckart acoustic streaming in a heptagonal chamber by multiple acoustic transducers

A new computational method was developed to simulate a two-dimensional Eckart acoustic streaming field in an ultrasonic heptagonal chamber actuated by multiple acoustic transducers with different associated frequencies and acoustic incident pressures. Simulation was conducted using the superposition of multiple spatial gradients of the Reynolds stresses and the second mean sound pressures at different frequencies. The developed method extends beyond the capabilities of the conventional method that is restricted to uniform frequency and incident pressure. Various acoustic streaming patterns can be feasibly generated by tuning the frequency and incident pressure of each individual transducer. The implementation of multiple acoustic transducers offers flexibility to control acoustic flows in microfluidic devices for various applications. Furthermore, the developed simulation method for acoustic streaming fields provides an optimization tool for the frequency, incident pressure and location of each transducer.     

引入漂移特性的用户兴趣模型优化研究

根据用户浏览网页时的操作行为,通过量化的方法建立起用户兴趣模型来反映用户兴趣,从而针对不同用户推荐其可能感兴趣的文章。基于兴趣模型的更新效率问题和用户兴趣的漂移特性,引入兴趣模型的时间分段机制和时间衰减机制,对兴趣模型进行了持续优化。实验表明,优化的兴趣模型在系统性能上有较大的提升,并能较好地反映出用户的兴趣变化,对于用户兴趣的表征更加准确,从而进一步提高了兴趣模型推荐文章的准确率。     

铝电解模糊控制系统的研究与应用

本文提出了一种自适应模糊控制方法．并应用在铝电解控制系统中。该方法将模糊控制函数的优化和模糊控制系统增益自适应调整集中于一体，可大大改善系统的品质和适应能力。仿真结论及应用表明．这种新型的模糊控制方法取得了良好的控制效果。     

利用混沌搜索全局最优的一种混合算法

把共轭梯度法与混沌优化方法相结合，提出了一种混合优化算法，该算法能使共轭梯度法跳出局部最优，最终获得全局最优，算法的收敛性也进行证明，仿真表明算法是有效的。     

基于P2P的结构化半分布式语义搜索算法

目前的搜索引擎算法大多是基于关键词搜索,体系上是集中式结构,在安全性和语义性上存在缺陷.基于P2P的结构化半分布式语义搜索算法--DPSS(Distributed P2P Semantic Search),通过研究词与词的关联度,在传统的P2P覆盖语义层,实现语义功能.同时,结合文章距离算法,能够应用在文本分类、文章聚类等领域.     

不同光谱与空间分辨率遥感图像融合方法的理论研究

多源遥感图像的融合处理是提高已获取同一目标影像的光谱和空间分辨率的重要手段,它可以增强多重数据分析处理能力,为大规模遥感应用研究提供一个良好的基础。对目前国内外现有的理论研究方法作了总结,将其分为像素级、特征级和决策级融合3大类型,分别进行了详细的阐述,同时引用具体研究成果对一些方法作了对比。最后,提出了该领域未来的主要研究与发展方向以供讨论。
     

宁夏地面气象观测记录电子报表管理系统的设计

为了提高地面气象观测记录电子报表的管理和使用,采用SQL Server、VC++、Java等技术在系统设计中应用。做了气象观测记录电子报表信息管理系统的总体设计和具体模块的实现。实验证明,系统运行良好,在气象数据服务方面起到良好作用。