Data leakage jeopardizes ecological applications of machine learning

被引：9

作者：

Stock, Andy ^{[1
]}

Gregr, Edward J. ^{[1
,2
]}

Chan, Kai M. A. ^{[1
]}

机构：

[1] Univ British Columbia, Inst Resources Environm & Sustainabil, Vancouver, BC, Canada

[2] SciTech Environm Consulting, Vancouver, BC, Canada

来源：

NATURE ECOLOGY & EVOLUTION | 2023年 / 7卷 / 11期

基金：

加拿大自然科学与工程研究理事会;

关键词：

VALIDATION;

D O I：

10.1038/s41559-023-02162-1

中图分类号：

Q14 [生态学（生物生态学）];

学科分类号：

071012 ; 0713 ;

摘要：

Machine learning is a popular tool in ecology but many scientific applications suffer from data leakage, causing misleading results. We highlight common pitfalls in ecological machine-learning methods and argue that discipline-specific model info sheets must be developed to aid in model evaluations.

引用

页码：1743 / 1745

页数：3

共 50 条

[1] Data leakage jeopardizes ecological applications of machine learning
Andy Stock
Edward J. Gregr
Kai M. A. Chan
Nature Ecology & Evolution, 2023, 7 : 1743 - 1745
[2] A review of supervised machine learning algorithms and their applications to ecological data
Crisci, C.
Ghattas, B.
Perera, G.
ECOLOGICAL MODELLING, 2012, 240 : 113 - 122
[3] Guiding questions to avoid data leakage in biological machine learning applications
Bernett, Judith
Grimm, Dominik G.
Haselbeck, Florian
Blumenthal, David B.
Joeres, Roman
Kalinina, Olga V.
List, Markus
NATURE METHODS, 2024, 21 (08) : 1444 - 1453
[4] ON DATA LEAKAGE PREVENTION AND MACHINE LEARNING
Domnik, Jan
Holland, Alexander
35TH BLED ECONFERENCE DIGITAL RESTRUCTURING AND HUMAN (RE)ACTION, BLED ECONFERENCE 2022, 2022, : 695 - 703
[5] Applications of symbolic machine learning to ecological modelling
Dzeroski, S
ECOLOGICAL MODELLING, 2001, 146 (1-3) : 263 - 273
[6] Machine learning of poorly predictable ecological data
Shan, Y.
Paull, D.
McKay, R. I.
ECOLOGICAL MODELLING, 2006, 195 (1-2) : 129 - 138
[7] Information Leakage from Data Updates in Machine Learning Models
Hui, Tian
Farokhi, Farhad
Ohrimenko, Olga
PROCEEDINGS OF THE 16TH ACM WORKSHOP ON ARTIFICIAL INTELLIGENCE AND SECURITY, AISEC 2023, 2023, : 35 - 41
[8] Editorial: Analysis and synthesis of ecological data by machine learning
Recknagel, Friedrich
Staiano, Antonino
ECOLOGICAL INFORMATICS, 2019, 53
[9] Machine Learning Applications for the Prediction of Bone Cement Leakage in Percutaneous Vertebroplasty
Li, Wenle
Wang, Jiaming
Liu, Wencai
Xu, Chan
Li, Wanying
Zhang, Kai
Su, Shibin
Li, Rong
Hu, Zhaohui
Liu, Qiang
Lu, Ruogu
Yin, Chengliang
FRONTIERS IN PUBLIC HEALTH, 2021, 9
[10] Special issue: Applications of machine learning to ecological modelling - Preface
Recknagel, F
ECOLOGICAL MODELLING, 2001, 146 (1-3) : 1 - 2

← 1 2 3 4 5 →