Mining of Multivariate Temporal Biological Data: A Framework for the Rational Design of Data-Driven Models.

Kamimura, R. T. ; Bicciato, S. ; Shimizu, H. ; Alford, J. ; Stephanopoulos, G. N.

Technical Information Center Oak Ridge Tennessee

关键词: Data management;    Information storage and retrieval systems;    Multivariate analysis;    Modeling;    Data collection;   

RP-ID  :  DE200415006266

学科分类：工程和技术（综合）

美国|英语

来源: National Technical Reports Library

【 摘 要 】

A framework is presented that emphasizes the need to understand the strengths and weaknesses of the data prior to modeling. In short, given a list of constraints, the idea is to let the data sort itself along those guidelines. Once the data has been organized into some coherent faction, the user has a better understanding of what the strengths and weaknesses of the data are as the analysis proceeds. The goal is to understand the character of the data so that the user is not overwhelmed but is able to systematically organize and decompose information so as to facilitate the analysis and build an effective model. The data analyzed is that from an industrial fermentation but the framework presented is generic enough that it can be used in any application involving multivariate time series data, such as time varying microarray measurements.