您的位置: 专家智库 > >

国家自然科学基金(06JJ2033)

作品数:2 被引量:12H指数:1
相关作者:罗飞更多>>
相关机构:湖南工业大学更多>>
发文基金:国家自然科学基金更多>>
相关领域:自动化与计算机技术更多>>

文献类型

  • 2篇中文期刊文章

领域

  • 2篇自动化与计算...

主题

  • 1篇应用性能
  • 1篇游戏
  • 1篇游戏规则
  • 1篇神经网
  • 1篇神经网络
  • 1篇数据挖掘
  • 1篇资源利用
  • 1篇网格
  • 1篇网格环境
  • 1篇混合数据
  • 1篇NN
  • 1篇RS
  • 1篇MAS
  • 1篇MODEL_...
  • 1篇粗糙集

机构

  • 1篇湖南工业大学

作者

  • 1篇罗飞

传媒

  • 1篇广西师范大学...
  • 1篇Scienc...

年份

  • 1篇2009
  • 1篇2007
2 条 记 录,以下是 1-2
排序方式:
Dynamic scheduling model of computing resource based on MAS cooperation mechanism被引量:11
2009年
Allocation of grid resources aims at improving resource utility and grid application performance. Currently, the algorithms proposed for this purpose do not fit well the autonomic, dynamic, distributive and heterogeneous features of the grid environment. According to MAS (multi-agent system) cooperation mechanism and market bidding game rules, a model of allocating allocation of grid resources based on market economy is introduced to reveal the relationship between supply and demand. This model can make good use of the studying and negotiating ability of consumers’ agent and takes full consideration of the consumer’s behavior, thus rendering the application and allocation of resource of the consumers rational and valid. In the meantime, the utility function of consumer is given; the existence and the uniqueness of Nash equilibrium point in the resource allocation game and the Nash equilibrium solution are discussed. A dynamic game algorithm of allocating grid resources is designed. Experimental results demonstrate that this algorithm diminishes effectively the unnecessary latency, improves significantly the smoothness of response time, the ratio of throughput and resource utility, thus rendering the supply and demand of the whole grid resource reasonable and the overall grid load balanceable.
JIANG WeiJin1,2, ZHANG LianMei3 & WANG Pu4 1 School of Computer and Electronic Engineering, Hunan University of Commerce, Changsha 410205, China
关键词:网格环境游戏规则应用性能资源利用
一种新的基于RS和NN的混合数据挖掘算法(英文)被引量:1
2007年
提出一种结合粗糙集理论和BP神经网络理论的新数据挖掘算法。算法利用粗糙集对属性的归约功能将数据仓库中的数据进行归约,将归约后的数据作为训练数据提供给神经网络。通过粗糙集归约,提高了训练数据表达的清晰度,也减少了神经网络的规模,同时利用神经网络又弥补了粗糙集对噪声数据敏感的不足。
罗飞
关键词:数据挖掘粗糙集神经网络
共1页<1>
聚类工具0