一种基于信息测度的科学数据集序列约减方法

计算物理 ›› 2009, Vol. 26 ›› Issue (6): 831-836.

一种基于信息测度的科学数据集序列约减方法

吴国清^1,2, 莫则尧², 陈虹²

1. 中国工程物理研究院研究生部, 北京 100088;
2. 北京应用物理与计算数学研究所, 北京 100094

收稿日期:2008-07-03 修回日期:2008-12-10 出版日期:2009-11-25 发布日期:2009-11-25
作者简介:吴国清(1980-),男,山东临沂,博士,主要从事科学数据挖掘、数据压缩研究,北京海淀区丰豪东路2号高性能计算中心100094.
基金资助:
国家自然科学基金(编号:90718029)资助项目

An Approach for Scientific Dataset Stream Reduction Based on Information Measures

WU Guoqing^1,2, MO Zeyao², CHEN Hong²

1. Graduate Department of China Academy of Engineering Physics, Beijing 100088, China;
2. Institute of Applied Physics and Computational Mathematics, Beijing 100094, China

Received:2008-07-03 Revised:2008-12-10 Online:2009-11-25 Published:2009-11-25

摘要/Abstract

摘要： 提出一种基于信息测度的数据集序列约减方法,研究如何从序列中抽样出具有较小相关性,同时不丢失具有重要物理特征的数据集.方法具有普适性,应用于激光与等离子体相互作用模拟程序的结果数据中,减少数据集间的相关性和信息冗余度,单个数据集的平均信息量较原数据集序列增加30%左右.

关键词: 科学数据, 数据约减, 仙农熵, 边际效用

Abstract: We propose a data reduction approach based on information theory. It comprises sampling of datssets based on mutual entropy and truncation based on offline Marginal Utility. The approach is a universal method for multi-dimensional scientific dataset streams. To show applicability, results obtained with plasma simulation data are presented, It reduces relationship and redundancy between datesets.

Key words: scientific datasets, data reduction, Shannon entropy, marginal utility

中图分类号:

O236

吴国清, 莫则尧, 陈虹. 一种基于信息测度的科学数据集序列约减方法[J]. 计算物理, 2009, 26(6): 831-836.

WU Guoqing, MO Zeyao, CHEN Hong. An Approach for Scientific Dataset Stream Reduction Based on Information Measures[J]. CHINESE JOURNAL OF COMPUTATIONAL PHYSICS, 2009, 26(6): 831-836.