Lime Amendment Efficacy Factors Analyzed with Gradient Boosting, 857 Observations

857 observations compiled to analyze factors influencing soil pH change after lime application. The dataset was used in a supervised machine learning framework with a Gradient Boosting Model (GBM) by author Floyd Muyembe Chipatela, last updated on 2026-04 22. The study achieved an R² of 89.7% and identified particle size, reaction time, and application rate as key factors.

Use Cases

• Predict soil pH response to lime amendments based on particle size, reaction time, and application rate.
• Evaluate the relative importance of factors like neutralizing value (NV) and limestone type for lime quality assessment.
• Train machine learning models to capture complex interactions in agricultural soil chemistry data.
• Optimize liming schedules based on the identified optimal response window of 90 to 400 days.

Strengths

• Model performance is quantitatively reported with an R² of 89.7%, low intercept, and RMSE.
• The analysis is based on 857 observations, providing a substantive sample size.
• Key influential factors and their interactions, such as particle size and reaction time, are explicitly identified.
• The study uses grouped cross-validation, suggesting a robust evaluation method.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is small at 49.6 KB, indicating limited scope.

Provenance

Source

figshare

Collection Method

Compiled from previous studies, as described in the methods section.

Time Range

Temporal coverage of the underlying observations is not specified.

Freshness

Last updated 2026-04-22 05:34:54; freshness should be verified.

Geography

Spatial coverage of the underlying soil samples is not specified.