Improving reliability of aggregation, numerical simulation and analysis of complex systems by empirical data

11th All-Russian Scientific and Practical Conference (with international participation) "Automation systems in education, science and production, 2017"

计算机科学;教育;自然科学

Dobronets, Boris S.^1 ; Popova, Olga A.^1

Institute of Space and Information Technology, Siberian Federal University, Kirenskogo 26, Krasnoyarsk

660074, Russia^1

关键词: Aggregate function;    Data representations;    Frequency distributions;    Functional dependency;    Important features;    Piecewise polynomial models;    Pre-processing stages;    Simulation and analysis;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/354/1/012006/pdf
DOI  :  10.1088/1757-899X/354/1/012006

来源: IOP

【 摘 要 】

The paper considers a new approach of regression modeling that uses aggregated data presented in the form of density functions. Approaches to Improving the reliability of aggregation of empirical data are considered: improving accuracy and estimating errors. We discuss the procedures of data aggregation as a preprocessing stage for subsequent to regression modeling. An important feature of study is demonstration of the way how represent the aggregated data. It is proposed to use piecewise polynomial models, including spline aggregate functions. We show that the proposed approach to data aggregation can be interpreted as the frequency distribution. To study its properties density function concept is used. Various types of mathematical models of data aggregation are discussed. For the construction of regression models, it is proposed to use data representation procedures based on piecewise polynomial models. New approaches to modeling functional dependencies based on spline aggregations are proposed.