基于B/S架构的数据在线可视化质量控制分析  

B/S Architecture Based Visual Data Quality Control and Analysis

作　　者：敏玉芳[1,2] 张耀南[1,2] 冯克庭[1,2] 赵雪茹[1,2] 康建芳[1,2] 

机构地区：[1]中国科学院旱区寒区环境与工程研究所,兰州730000 [2]甘肃高性能网格计算中心,兰州730000

出　　处：《科研信息化技术与应用》2011年第5期26-35,共10页E-science Technology & Application

基　　金：中国科学院信息化专项项目"冰雪冻土环境本底与可持续发展专题数据库""特殊学科人才培养冰川冻土学科点(J0930003/J0109)"

摘　　要：针对基于Web的科学数据共享平台中时间序列的科学数据,提出了基于B/S架构的两阶段数据质量控制的分析方法:(1)数据入库前常规质量控制,包括初步的数据整理,实现结构统一、格式一致的观测资料入库管理,完成数据一致性检验等;(2)基于可视化的在线数据质量控制。根据数据请求,在线提供极值检验法、3δ检验法、Dixon准则、Grubbs准则异常值检测方法,建立可视化图表展示的数据质量控制方法。该方法可支持数据核对和数据异常点剔除,使用户直观形象地了解数据整体、掌握时间变化规律及数据间的内在联系。According to the scientific data of time sequence in the scientific data platform based on Web, we proposed a B/S based system where we divided the process of data quality control analysis into two parts. One is data quality control by conventional methods before data inventory, including data pre-processing and realization of the observed data structure unity and format consistent. The other is B/S architecture based visual data quality control and analysis. According to data request, the system provides online outlier detection methods, including 3（~＇test method, Dixon＇ test method, and Grubbs＇ test method, etc. After quality control, the charts will be fed back to users. The system will provide convenience for data detection and data abnormal points out methods, and this will help users to understand the overall of data and grasp the change rules over time and the intrinsic relationship between the data.

关 键 词：质量控制 B/S 可视化 数据分析 科学数据 Amcharts 

分 类 号：O211.61[理学—概率论与数理统计]