摘要
对海量图书关键词特征进行准确的检索与定位,可以提高图书管理的质量和效率。进行关键词特征检索定位时,需要依据图书关键词特征的关联性,对图书关键词特征检测,而传统的蚁群规则算法是对图书关键词特征信息浓度策略进行更新完成检索定位,难以建立图书关键词特征之间的关联,不能对图书关键词特征进行准确挖掘,降低了海量图书关键词特征检索定位的精度。提出采用多因素方差的海量图书关键词特征检索定位优化方法,先利用多因素方差得到图书关键词特征挖掘规律,结合蚁群算法,根据蚁群适应度概率正则训练迁移法则,得到种群变异当前时刻的图书检索关键词特征,结合用户选择特定图书文献进行检索的概率,建立海量图书关键词特征检索模型和海量图书关键词特征的定位模型;实现海量图书关键词特征的快速检索定位。仿真结果表明,与传统方法相比,所提算法进行海量图书关键词特征检索定位时,具有较高的精度和效率。
A retrieval and location optimization method of keyword feature in massive books is proposed using multifactor variance. Firstly the muhifactor variance is used to obtain the rule of feature mining. Then the keyword feature of book retrieval in current population variation is obtained according to the ant colony fitness probability regular training migration law integrated with ant colony algorithm. Finally the retrieval and location model of keyword feature in massive books is built and integrated with probability of books and documents retrieval specifically selected by user to achieve fast retrieval and location. The simulation results show that the algorithm mentioned above has higher precision and efficiency than traditional methods.
出处
《计算机仿真》
CSCD
北大核心
2016年第9期422-425,共4页
Computer Simulation
关键词
图书
关键词
特征
检索定位
Books
Key words
Feature
Retrieve and location