基于互信息下粒子群优化的属性约简算法

发布时间：2018-04-30 23:15

本文选题：互信息 + 粒子群优化　；参考：《电子学报》2017年11期

【摘要】：最小属性约简是粗糙集理论中属性约简的优化问题.在寻找最小属性约简的问题上,基于粒子群优化的属性约简算法(ARPSO算法)优于传统的属性约简算法.在现有的ARPSO算法中,正域部分通常被作为启发式信息,但是它并不能够很好地衡量不确定性,而互信息是粗糙集理论中一种更有效的度量不确定信息的重要工具.为此,提出基于互信息下的粒子群优化的属性约简算法(MIPSO算法),该算法把互信息作为适应度函数,通过增强粒子能迅速靠近吸引子的这一特性,改进了内嵌区域震荡搜索的粒子群优化算法(简记为RSPSO算法),防止算法较早的陷入局部最优,使得粒子群中的粒子更快的找到最优值,因此使得算法尽可能实现全局收敛.实验结果表明,该算法不仅提高了寻优的能力,加快了算法的速度,提升了算法的精度,而且也能够使得约简后剩余属性的互信息值与约简前所有属性的互信息值近似相等.
[Abstract]:Minimum attribute reduction is an optimization problem of attribute reduction in rough set theory. In the problem of finding minimum attribute reduction, the attribute reduction algorithm based on particle swarm optimization (PSO) is superior to the traditional attribute reduction algorithm. In existing ARPSO algorithms, the positive domain is usually used as heuristic information, but it can not measure uncertainty well, and mutual information is an important tool to measure uncertainty in rough set theory. Therefore, an attribute reduction algorithm based on mutual information for particle swarm optimization (PSO) is proposed. This algorithm regards mutual information as a fitness function and enhances the property that particles can quickly approach the attractor. This paper improves the particle swarm optimization algorithm for embedded region oscillation search (abbreviated as RSPSO algorithm) to prevent the algorithm from falling into local optimum earlier, so that the particles in the particle swarm can find the optimal value more quickly, so that the algorithm can achieve global convergence as much as possible. The experimental results show that the algorithm not only improves the optimization ability, accelerates the speed of the algorithm and improves the accuracy of the algorithm, but also makes the mutual information value of the remaining attributes after reduction approximately equal to the mutual information values of all the attributes before reduction.
【作者单位】：太原理工大学;太原理工大学国际教育交流学院;
【基金】：人社部留学回国人员科技活动择优资助项目(No.2013-68) 山西省自然科学基金(No.2014011018-2) 山西省回国留学人员科研资助项目No.2013-033,No.2015-045)
【分类号】：TP18

【相似文献】