摘要
针对传统关键数据提取模型提取准确率和召回率双低的问题,建立了一个新的船舶航线数据库关键数据提取数学模型。该模型首先需要构建一个分类器,利用分类器将数据库中的数据进行分类;然后将四叉树、R-树相结合构建一个复合型OR-树结构关键数据索引;最后利用建立好的关键数据索引在分类好的数据中提取关键数据。结果表明:与利用主成分分析或粗糙集理论建立的提取模型相比,利用本次构建的模型进行关键数据提取,提取准确率提高3.3%和9.4%,召回率提高6.2%和7.9%。
Aiming at the problem of low accuracy and recall rate of traditional key data extraction model, a new mathematical model for key data extraction from ship route database is established. The model first needs to construct a classifier to classify the data in the database by using the classifier;then it combines quadtree and R-tree to construct a composite ORtree structure key data index;finally, it uses the established key data index to extract key data from the classified data. The results show that compared with the extraction model based on principal component analysis or rough set theory, the extraction accuracy and recall rate of this model are improved by 3.3% and 9.4% and 6.2% and 7.9% respectively.
作者
谢慧杰
XIE Hui-jie(Xuanhua Science and Technology Vocational College,Zhangjiakou 075100,China)
出处
《舰船科学技术》
北大核心
2019年第6期52-54,共3页
Ship Science and Technology
关键词
航线数据库
关键数据
提取数学模型
分类
数据索引
route database
key data
extraction of mathematical models
classification
data index
