多源异构数据的多维决策分析与可视化方法

随着各行业各业务信息系统的不断建设与使用,行业积累了越来越多的多源异构的历史数据,快速发现有价值信息已成为一个主要的研究方向.以气象数据为例,针对气象历史数据的海量多源异构特点,结合气象业务多维决策分析与可视化的具体需求,提出一种面向多源异构数据的多维决策分析与可视化方法,该方法实现了多源异构数据的自动抽取与转换,多用户的在线多维决策分析模型设计以及分析结果的直观多形式可视化技术.通过该方法研制的系统在嘉定气象局业务中得到成功应用,有效提升了对气象数据的理解和利用水平;同时对其他行业多源异构数据的处理具有很好的借鉴作用.

A multi-dimensional decision-making analysis and visualization method for solving multi-source and heterogeneous datasets

With the rapid development and application of all kind of industry information systems,it has been a very important research area that how to effectively exploit valuable knowledge from multi-source and heterogeneous.Taking meteorological datasets as an example,a multi-dimensional decision-making analysis and visualization method for solving multi-source and heterogeneous datasets is proposed in this paper according to the features of the numerous multi-source heterogeneous meteorological historical datasels and the specific requirement of meteorological decision-making analysis and visualization.Three main process of our method are detailed as follow：auto-extraction and format transformation of original datasels,many users＇ online multi-dimensional decision-making analysis model,and multi-type intuitional visualizations of result mined.Finally,the system has been deployed in Jiading Meteorological Administration and is helpful to improve the ability of understanding meteorological datasets.As the same time,our method will be adopted by the other similar industry in the future.