Novel deep learning algorithm in soil erodibility factor predicting at a continental scale
Bibliographic record
Abstract
Soil erosion poses significant environmental and economic challenges, adversely affecting soil fertility and global agricultural productivity. We developed a novel model based on the Multi-Head Squeeze-and-Excitation Residual One-Dimensional Convolutional Neural Network (MH-SE-Res1DNet) to predict the soil erodibility factor ( K ) across Europe, representing the first application of this model for such a purpose worldwide. We conducted a comparative analysis using five benchmark machine learning algorithms, i.e., Random Forest (RF), Artificial Neural Network–Multilayer Perceptron (ANN-MLP), Support Vector Regression (SVR), Alternating Model Tree (AMT), and Pace Regression (PR), to assess the efficacy of our model. The results showed that the MH-SE-Res1DNet deep learning model had an outstanding ability for the K prediction. The model's lowest error (MAE=0.0025, RMSE=0.0031) and highest coefficient of determination (R 2 =0.943) were attained during the validation phase. Benchmark models demonstrated lower performance compared to the MH-SE-Res1DNet model, with R 2 values ranging from 0.880 to 0.912 and slightly higher errors across MAE and RMSE metrics. The sensitivity analysis of MH-SE-Res1DNet showed that its performance depends predominantly on key soil factors, particularly topsoil texture (M) and organic matter (OM) concentration. This model establishes a data-driven framework that significantly advances soil erodibility prediction by leveraging machine learning. It surpasses traditional methods and existing machine learning approaches in accuracy, efficiency, and scalability, setting a new benchmark for soil conservation planning and enabling adaptable, evidence-based land management strategies across Europe and worldwide.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".