大型数据库中需求知识资源有效检测仿真研究

摘要

In this article,we propose a detection method for requirement knowledge resource in large database based on singular decomposition of maximal Lyapunov index to achieve efficient mining for requirement resource data.First of all,we carry out space reconstruction of high dimensional phase and QR decomposition for information flow of resource data to establish mesh distribution matrix of Lyapunove exponential spectrum of requirement knowledge resource,and then decompose feature vector matrix of requirement knowledge resource by using decomposition theory of singular value,thus,carry out anti-jamming filtering processing for feature vector of requirement knowledge resource.At last,we conduct fusion cluster analysis for resource feature vector and achieve optimization detection of requirement knowledge resource.The simulation results demonstrate that the method in this paper has short execution time for resource detection of requirement knowledge and high mining accuracy.%大型数据库中需求知识资源的有效检测,能够提高数据库资源的利用率.对需求知识资源的有效检测,需要对资源特征向量进行抗干扰滤波处理,实现融合聚类分析,完成对大型数据库资源的有效检测.传统方法依据层次聚类中的凝聚思想,对资源位置进行管理产生聚类,但忽略了对资源特征向量进行融合聚类分析,导致检测精度偏低.提出基于最大Lyapunove指数奇异分解的大型数据库中需求知识资源检测方法,实现需求资源数据高效挖掘.首先对资源数据信息流进行高维相空间重构和QR分解,组建需求知识资源的Lyapunove指数谱的网格分布矩阵,利用奇异值分解理论对需求知识资源特征向量矩阵行分解,对需求知识资源特征向量进行抗干扰滤波处理,对资源特征向量进行融合聚类分析,实现需求知识资源的优化检测.实验结果表明,所提方法进行需求知识资源检测的执行时间较短、挖掘精度较优.