摘要
针对具有行为规律的目标搜索问题,提出一种搜索资源分配算法.该方法以目标在各搜索区域的概率分布为基础,利用最优搜索理论分配区域搜索时长;以目标在各搜索区域的行为规律为基础,利用包络检测等方法决定区域开始搜索的时刻.针对网站关键词搜索的实验显示,根据目标行为规律在时间上相关程度的不同,本算法相对于最优搜索算法的性能提升在15%~50%之间,在对大量信息源进行信息搜索时具有应用价值.
This paper proposes a new algorithm for search resources allocation to search targets with behavior rule. In the new algorithm, the optimal search theory is used to allocate the search periods for searchzones based on the probability distribution of the targets, and a target-behavior based envelope detectionalgorithm is developed to decide the time instants when the search begins. The experiment results of keywords searching on network sites indicate that the proposed algorithm has much better performance than theoptimal search method, and the performance gain is between 15% and 50% for different correlations of behavior rules. The algorithm will find application in searching mass information resources.
出处
《电讯技术》
北大核心
2014年第2期195-200,共6页
Telecommunication Engineering
关键词
资源分配
最优搜索
行为规律
目标搜索
resources allocation
optimal search
behavior rule
target search
