Big data analysis via model reduction methods

Zabielin, Stanislav

Big data analysis via model reduction methods

Файли

109517-299050-1-10-20180727.pdf (130.87 KB)

Дата

2018

Автори

Zabielin, Stanislav

Видавець

КПІ ім. Ігоря Сікорського

Анотація

The enormous growth in the size of data has been observed in recent years being a key factor of the Big Data scenario. Big Data require a new high-performance processing. The use of big data preprocessing methods for data mining in big data is reviewed in this paper. The definition, attributes and categorization of data preprocessing approaches in big data are introduced. The relation between big data and data preprocessing throughout all families of methods and advanced data technologies are likewise analyzed. Furthermore, research challenges are discussed, while concentrating on improvements in certain families of data preprocessing methods and applications based on new big data learning paradigms.

Ключові слова

nonlinear mapping, dimension reduction, big data, modelling, non-linear dynamic objects, dimensional reduction, diffusion maps, kernel method of main components

Бібліографічний опис

Zabielin, S. Big data analysis via model reduction methods / Stanislav Zabielin // Системні дослідження та інформаційні технології : міжнародний науково-технічний журнал. – 2018. – № 2. – С. 35-41. – Бібліогр.: 22 назви.