ACTIVE LEARNING METHODS FOR LABELLING DATASETS
Multidisciplinary Journal of Research in Engineering and Technology · 15.01.2023 · 10 (1) · 25-32
DOI: 10.65521/mjret.v10i1.1167 ↗Kurz gefasst
Fachartikel in der Zeitschrift Multidisciplinary Journal of Research in Engineering and Technology, online erschienen am 15.01.2023. Verfasst von Atharv Wani und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Data scientists are challenged with more data than they will ever be able to analyse as data collection and storage costs continue to drop. The most fascinating developments in machine learning require vast amounts of data. But it also creates a new challenge for the machine learning community, as all supervised learning-based machine learning applications remain practically useless without labelled data. Labelling large datasets has become a vital challenge. A specific instance of Supervised Machine Learning is Active Learning. By actively choosing the important data points, this method builds a highperformance classifier thus minimizing the size of the training dataset.
Zitieren
APA
Wani, A., Chowdary, G. C. S., Meleet, M. M., & R, A. M. (2023). ACTIVE LEARNING METHODS FOR LABELLING DATASETS. Multidisciplinary Journal of Research in Engineering and Technology, 10(1), 25-32. https://doi.org/10.65521/mjret.v10i1.1167
Vancouver
Wani A, Chowdary GCS, Meleet MM, R AM. ACTIVE LEARNING METHODS FOR LABELLING DATASETS. Multidisciplinary Journal of Research in Engineering and Technology. 2023;10(1):25-32. doi:10.65521/mjret.v10i1.1167
BibTeX
@article{Wani2023,
author = {Wani, Atharv and Chowdary, Gorla Charan Sai and Meleet, Merin Meleet and R, Anala M},
title = {ACTIVE LEARNING METHODS FOR LABELLING DATASETS},
journal = {Multidisciplinary Journal of Research in Engineering and Technology},
year = {2023},
volume = {10},
number = {1},
pages = {25--32},
doi = {10.65521/mjret.v10i1.1167}
}
EUVIMED