欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
国家自然科学基金委设立“重点学术期刊专项基金”资助国内优秀学术期刊，在优秀期刊的遴选和资助效果评价过程中，存在期刊数量大，评价指标多等问题．基于此，提出一种针对大规模高维数据的简化的区间数据主成分分析方法（simplified principal component analysis，SPCA）．该方法将区间主成分分析分解成2个基本阶段：第一是如何更加简单和高精度地计算数据集合的主轴，第二是如何绘制可视性与可解释性都更强的主平面图，以增强研究人员对大规模数据主要特征的洞察能力．采用SPCA方法，一方面能够从整体上研究各学科期刊的差异，说明学科之间的不可比较性；另一方面能够筛选出衡量期刊水平的关键指标，遴选优秀期刊，同时分析连续受资助期刊的动态资助效果．
英文摘要：
National Natural Science Foundation of China （NSFC） encountered many problems, such as an increasing number of applicants and a great many indices, when selecting excellent academic journals as grantees of ＂Key Academic Journal Fund＂ and evaluating funding effects. This paper puts forward a method for Principal Component Analysis on interval data, referred to as Simplified Principal Component Analysis （SPCA） , specific for large-scaled and high-dimensional data. The proposed method involves two steps, first look for factor axes of the original large-scaled data and then illustrate the analytical results on a low-dimensional space with strengthened visibility and high interpretation. Performing SPCA on academic journal dataset helps to understand the differences in all disciplines and to choose key criteria for selecting excellent journals and measuring funding effect dynamically.
同期刊论文项目
同项目期刊论文
Simulation of exit choosing in pedestrian evacuation with consideration of the direction visual fiel
A new model for studying the SO-based pre-trip information release strategy and route choice behavio
A cellular automaton model of traffic considering the dynamic evolution of velocity randomization pr
Continuum modeling of park-and-ride services in a linear monocentric city with deterministic mode ch
Route choice in pedestrian evacuation under conditions of good and zero visibility: Experimental and
期刊信息
位置：