Disturbed-entropy: A simple data quality assessment approach
Özet
From the perspective of information value, we proposed a simple and effective approach to assess data quality, called disturbed-entropy. In specific, considering image classification task, the existing samples per category are statistically represented as a pixel prototype, which is used to disturb the unseen samples. Then, the entropy of disturbed image is calculated based on predicted probability. Both the numerical and visual experiments are conducted to show the effect. In case of same data budget, the performance comparison based on selected good and bad data is significant and consistent. This work attempts to gain insight into data quality and redundancy. (C) 2022 The Author(s). Published by Elsevier B.V. on behalf of The Korean Institute of Communications and Information Sciences.
Anahtar kelimeler
Atıf için künye
Li, Y., Chao, X., & Ercişli, S. (2022). Disturbed-entropy: A simple data quality assessment approach. ICT EXPRESS, 8(3), 309–312. https://doi.org/10.1016/j.icte.2022.01.006
BibTeX
@article{li2022-264979,
author = {Li, Yang and Chao, Xuewei and Ercişli, Sezai},
title = {Disturbed-entropy: A simple data quality assessment approach},
journal = {ICT EXPRESS},
year = {2022},
volume = {8},
number = {3},
pages = {309--312},
issn = {2405-9595},
doi = {10.1016/j.icte.2022.01.006},
url = {https://akademik.atauni.edu.tr/yayinlar/264979-disturbed-entropy-a-simple-data-quality-assessment-approach}
}