The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping
Data leakage occurs when there is an overlap between the data used for model fitting and hyperparameter tuning, and those used for testing. This overlap biases the model performance, making it uninformative regarding the model’s ability to generalize. This is a significant issue in machine learning...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Geoderma |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0016706125000618 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850228052105101312 |
|---|---|
| author | Kingsley John Daniel D. Saurette Brandon Heung |
| author_facet | Kingsley John Daniel D. Saurette Brandon Heung |
| author_sort | Kingsley John |
| collection | DOAJ |
| description | Data leakage occurs when there is an overlap between the data used for model fitting and hyperparameter tuning, and those used for testing. This overlap biases the model performance, making it uninformative regarding the model’s ability to generalize. This is a significant issue in machine learning and predictive soil mapping, compromising model reliability. To demonstrate this issue, the 3-dimensional (3D) digital soil mapping (DSM) approach, whereby depth is used as a predictor of soil properties, was investigated. We compare two common approaches from the literature: leave-sample-out cross-validation (LSOCV) versus leave-profile-out cross-validation (LPOCV). Here, we argue that LSOCV results in contamination of the test dataset due to the potential vertical autocorrelation of soil properties from different samples within the same profile, and a more appropriate approach for testing 3D DSM models should be to fully partition all soil samples from the same profile to either the training or test dataset (i.e., LPOCV). Using the Ottawa region of Ontario, Canada, as a case study, cation exchange capacity (CEC), clay content, pH, and total organic carbon (TOC) were predicted using machine learning, and the discrepancy in accuracy metrics was reported. Furthermore, we evaluated the effects of data augmentation (i.e., the creation of additional synthetic data points from the original data) on accuracy metrics, a common practice in 3D DSM. Here, it was shown that with the augmented dataset, LSOCV generated overly optimistic accuracy metrics (e.g., CCC) that were 29–62% higher than LPOCV, while for the non-augmented data, the accuracy metrics were 8–18% higher, suggesting that vertical autocorrelation had a strong influence on inflating model accuracy through data leakage. As such, we strongly urge DSM practitioners to provide greater clarity when describing how model accuracy metrics were ascertained and to consider the use of LPOCV when applied to 3D DSM. This brings about broader concerns that policymakers and stakeholders may use map products with the false impression that the maps are more accurate than they are. Future research should focus on refining DSM methods and considering data structure to prevent data leakage in modelling soil properties. |
| format | Article |
| id | doaj-art-3694dfca08d949f984a58dec9e037eb1 |
| institution | OA Journals |
| issn | 1872-6259 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Geoderma |
| spelling | doaj-art-3694dfca08d949f984a58dec9e037eb12025-08-20T02:04:38ZengElsevierGeoderma1872-62592025-03-0145511722310.1016/j.geoderma.2025.117223The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mappingKingsley John0Daniel D. Saurette1Brandon Heung2Department of Plant, Food, and Environmental Sciences, Faculty of Agriculture, Dalhousie University, PO Box 550, 21 Cox Rd., Truro, NS B2N 5E3, CanadaOntario Ministry of Agriculture, Food and Agribusiness, 1 Stone Rd W, Guelph, ON N1G 4Y2, CanadaDepartment of Plant, Food, and Environmental Sciences, Faculty of Agriculture, Dalhousie University, PO Box 550, 21 Cox Rd., Truro, NS B2N 5E3, Canada; Corresponding author.Data leakage occurs when there is an overlap between the data used for model fitting and hyperparameter tuning, and those used for testing. This overlap biases the model performance, making it uninformative regarding the model’s ability to generalize. This is a significant issue in machine learning and predictive soil mapping, compromising model reliability. To demonstrate this issue, the 3-dimensional (3D) digital soil mapping (DSM) approach, whereby depth is used as a predictor of soil properties, was investigated. We compare two common approaches from the literature: leave-sample-out cross-validation (LSOCV) versus leave-profile-out cross-validation (LPOCV). Here, we argue that LSOCV results in contamination of the test dataset due to the potential vertical autocorrelation of soil properties from different samples within the same profile, and a more appropriate approach for testing 3D DSM models should be to fully partition all soil samples from the same profile to either the training or test dataset (i.e., LPOCV). Using the Ottawa region of Ontario, Canada, as a case study, cation exchange capacity (CEC), clay content, pH, and total organic carbon (TOC) were predicted using machine learning, and the discrepancy in accuracy metrics was reported. Furthermore, we evaluated the effects of data augmentation (i.e., the creation of additional synthetic data points from the original data) on accuracy metrics, a common practice in 3D DSM. Here, it was shown that with the augmented dataset, LSOCV generated overly optimistic accuracy metrics (e.g., CCC) that were 29–62% higher than LPOCV, while for the non-augmented data, the accuracy metrics were 8–18% higher, suggesting that vertical autocorrelation had a strong influence on inflating model accuracy through data leakage. As such, we strongly urge DSM practitioners to provide greater clarity when describing how model accuracy metrics were ascertained and to consider the use of LPOCV when applied to 3D DSM. This brings about broader concerns that policymakers and stakeholders may use map products with the false impression that the maps are more accurate than they are. Future research should focus on refining DSM methods and considering data structure to prevent data leakage in modelling soil properties.http://www.sciencedirect.com/science/article/pii/S0016706125000618AccuracyMachine learningPedometricsSoil propertiesVertical autocorrelation |
| spellingShingle | Kingsley John Daniel D. Saurette Brandon Heung The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping Geoderma Accuracy Machine learning Pedometrics Soil properties Vertical autocorrelation |
| title | The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping |
| title_full | The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping |
| title_fullStr | The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping |
| title_full_unstemmed | The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping |
| title_short | The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping |
| title_sort | problematic case of data leakage a case for leave profile out cross validation in 3 dimensional digital soil mapping |
| topic | Accuracy Machine learning Pedometrics Soil properties Vertical autocorrelation |
| url | http://www.sciencedirect.com/science/article/pii/S0016706125000618 |
| work_keys_str_mv | AT kingsleyjohn theproblematiccaseofdataleakageacaseforleaveprofileoutcrossvalidationin3dimensionaldigitalsoilmapping AT danieldsaurette theproblematiccaseofdataleakageacaseforleaveprofileoutcrossvalidationin3dimensionaldigitalsoilmapping AT brandonheung theproblematiccaseofdataleakageacaseforleaveprofileoutcrossvalidationin3dimensionaldigitalsoilmapping AT kingsleyjohn problematiccaseofdataleakageacaseforleaveprofileoutcrossvalidationin3dimensionaldigitalsoilmapping AT danieldsaurette problematiccaseofdataleakageacaseforleaveprofileoutcrossvalidationin3dimensionaldigitalsoilmapping AT brandonheung problematiccaseofdataleakageacaseforleaveprofileoutcrossvalidationin3dimensionaldigitalsoilmapping |