时间序列搜索问题的在线决策算法  

Online Decision Algorithm for Time Series Search Problem

作　　者：张瑞英[1] 张远平[1] 

机构地区：[1]兰州理工大学计算机与通信学院,中国兰州730050

出　　处：《湖南师范大学自然科学学报》2011年第3期15-19,共5页Journal of Natural Science of Hunan Normal University

基　　金：动态Holonic制造系统建模及实时调度策略研究资助项目(61064011)

摘　　要：时间序列搜索是现实中一个典型的在线交易决策问题,已有各种不同的模型用确定性或随机性算法得到求解.基于引入利润函数的模型基础,提出了随机性算法,分析了其竞争比,并通过实例表明,该算法与确定性算法相比较在一定情况下可以有效地降低算法竞争比.Time series search which has the massive applications is a typical online trading decision problem.A variety of solutions have been given to different models by using deterministic or random algorithms.Based on the model with profit function,a random algorithm is proposed and its competitive ratio is analyzed.The examples demonstrate that this random algorithm can effectively reduce the competitive ratio in some cases comparing with the deterministic one.

关 键 词：时间序列搜索 在线交易算法 确定性算法 随机性算法 竞争比 

分 类 号：TP302[自动化与计算机技术—计算机系统结构]