摘要
分析原始基金数据的特点,依据知识发现的基本原理,对科研基金数据进行拆分、清洗、分类与甄别,建立起基金分类体系,再利用该体系对科研基金数据进行标注。通过"聚类-完善分类体系-标注"的循环,完成对科研基金数据的加工。
The paper analyzes the characteristics of original scientific research fund data, according to the basic principles of know edge discovery, it carries out partitioning, cleaning, classification and screening on scientific research fund data, constructs fund classification system, then uses the system to mark the scientific research fund data. Through " clustering - classification system perfection - marking " cycle to complete the scientific research fund data processing.
出处
《医学信息学杂志》
CAS
2014年第6期38-43,共6页
Journal of Medical Informatics