A Storage and Classification Algorithm for Concept Drift Data Streams Based on OS-ELM

Ying Mei (), Chengbo Lu (), Chenghao Tang (), Yonghua Wu () and Guoqing Meng ()
Additional contact information
Ying Mei: Lishui University
Chengbo Lu: Lishui University
Chenghao Tang: Hangzhou Steam Turbine Co., Ltd.
Yonghua Wu: Zhejiang Sci-Tech University
Guoqing Meng: Zhejiang Sci-Tech University

Journal of Classification, 2025, vol. 42, issue 2, No 7, 414-434

Abstract: Abstract Many methods can deal with some special cases of data streams (e.g., concept drift) currently; however, these methods need to store historical data and access them repeatedly, which is inconsistent with the single-channel characteristics of data streams, and it requires a large amount of memory space to save data when very slow gradual drift occurs due to the infinite feature of data streams. To address this problem, this paper proposes a concept drift data stream storage and classification algorithm based on OS-ELM (SC-OS-ELM), which uses a matrix of fixed size to save the feature information of historical data, and retrains the classifier by accessing this feature matrix when needed. It achieves the goal of improving the classification accuracy while storing the feature information of historical data in a small and constant memory space. It can help algorithms such as the DDM to solve the data storage problem, making it more applicable. Comparative experimental results on 15 artificial and real data streams validate the effectiveness of SC-OS-ELM, with a significant reduction in the amount of memory space required.

Keywords: Concept drift; Data stream; OS-ELM; Classification; Storage (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s00357-024-09499-7 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:jclass:v:42:y:2025:i:2:d:10.1007_s00357-024-09499-7

Ordering information: This journal article can be ordered from
http://www.springer. ... hods/journal/357/PS2

DOI: 10.1007/s00357-024-09499-7

Access Statistics for this article

Journal of Classification is currently edited by Douglas Steinley

More articles in Journal of Classification from Springer, The Classification Society
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().