最佳任务调度下的Web数据优化聚类算法

Web Data Optimization Clustering Algorithm Under the Optimal Task Scheduling

在对最佳任务调度下的Web数据进行优化聚类的过程中,容易出现原始数据损失的情况,导致传统数据优化聚类算法,由于忽略初始数据,无法有效实现Web数据优化聚类。提出一种基于粒子群优化的最佳任务调度下Web数据优化聚类算法,依据任务价值密度以及执行紧迫性,塑造动态优先级,通过适应度函数对分类计划进行评价,给出类间距与类内距计算公式,对相关参数和各粒子的位置以及速度向量进行初始化操作;求出粒子的适应度;求出粒子个体最优与群最优;依据粒子群优化算法的位置以及速度对当前位置和速度进行更新;通过K-means算法对EHCF进行聚类,直至全部Web数据聚类完成。仿真实验结果表明,所提方法在Web数据优化聚类上具有很高的优越性。

Under the optimal task scheduling of Web data to optimize the process of clustering, prone to raw data loss situation, lead to traditional data optimized clustering algorithm, due to ignore the initial data, cannot be effectively optimized Web data clustering, put forward a kind of optimal task scheduling based on particle swarm optimization under the Web data optimized clustering algorithm, based on the density of task value and implement the urgency, build dynamic priority,through fitness function to evaluate the classification scheme, gives calculation formula and class from the class spacing,the related parameters and initialized each particle＇s position and velocity vector operation; The fitness of particle; Calculated particle optimal individual and group of optimal; According to the position of the particle swarm optimization algorithm,and the speed to update the current position and speed; Through the K- means algorithm for clustering EHCF, until all completed Web data clustering. The simulation results show that the proposed method on the Web data optimization clustering has the advantages of high.