Piecewise Polynomial Aggregation as Preprocessing for Data Numerical Modeling
URI (для ссылок/цитирований):
http://iopscience.iop.org/article/10.1088/1742-6596/1015/3/032028/metahttps://elib.sfu-kras.ru/handle/2311/110683
Автор:
B, S Dobronets
O, A Popova
Коллективный автор:
Институт космических и информационных технологий
Кафедра систем искусственного интеллекта
Дата:
2018Журнал:
IOP Conf. Series: Journal of Physics: Conf. Series 1015 (2018)Квартиль журнала в Scopus:
Q3Квартиль журнала в Web of Science:
без квартиляБиблиографическое описание:
B, S Dobronets. Piecewise Polynomial Aggregation as Preprocessing for Data Numerical Modeling [Текст] / S Dobronets B, A Popova O // IOP Conf. Series: Journal of Physics: Conf. Series 1015 (2018): Journal of Physics: Conf. Series. — 2018. — Т. 032028.Аннотация:
Data aggregation issues for numerical modeling are reviewed in the present study.
The authors discuss data aggregation procedures as preprocessing for subsequent numerical
modeling. To calculate the data aggregation, the authors propose using numerical probabilistic
analysis (NPA). An important feature of this study is how the authors represent the aggregated
data. The study shows that the offered approach to data aggregation can be interpreted as the
frequency distribution of a variable. To study its properties, the density function is used. For
this purpose, the authors propose using the piecewise polynomial models. A suitable example
of such approach is the spline. The authors show that their approach to data aggregation allows
reducing the level of data uncertainty and significantly increasing the efficiency of numerical
calculations. To demonstrate the degree of the correspondence of the proposed methods to
reality, the authors developed a theoretical framework and considered numerical examples
devoted to time series aggregation