Coffee Leaf Rust Incidence Predictions from 9,850 Kenyan Farm Plot Observations

9,850 plot-level observations from six major coffee-producing counties in Kenya form the basis for this comparative model evaluation. The dataset, created by Maurice Wanyonyi and last updated in April 2026, compares Bayesian hierarchical inference and supervised machine learning approaches for predicting coffee rust incidence. It focuses on microclimatic moisture variables like leaf wetness duration and relative humidity as key predictors.

Use Cases

• Benchmarking logistic regression against ensemble methods for agricultural disease classification based on the described model comparison.
• Analyzing nonlinear threshold effects of humidity and leaf wetness on infection probability based on the partial dependence and SHAP analyses mentioned.
• Quantifying prediction uncertainty and spatial transmission dynamics in crop disease models based on the described Bayesian hierarchical framework.
• Developing decision-support tools for precision disease management in tropical coffee systems based on the described scalable framework.

Strengths

• Contains 9,850 plot-level observations providing a substantial basis for model training and evaluation.
• Models were evaluated across six major coffee-producing counties in Kenya, suggesting geographic diversity.
• The logistic regression model achieved a high discriminative performance (AUC ROC = 0.867), as reported in the description.
• Data is shared under a permissive CC-BY-4.0 license, facilitating reuse.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for specific modeling tasks.
• The dataset is very small at 5.5 KB, indicating limited raw data or a summary of model outputs.

Provenance

Source

Maurice Wanyonyi via figshare.

Collection Method

Longitudinal data collection from coffee farms, as described in the study.

Time Range

null

Freshness

Last updated 2026-04-21 17:43:26; freshness should be verified.

Geography

Six major coffee-producing counties in Kenya.