Model Performance Metrics Across Four Locations with Split Ratio Comparison

Usman Ali published a dataset on figshare in April 2026 comparing model performance across four locations. The dataset, stored in a 9.5 KB XLS file, presents R-squared and Mean Absolute Error metrics as mean ± standard deviation for different train-test split ratios. The standard deviation indicates the internal consistency of predictions within each split ratio.

Use Cases

• Compare model stability across different data split ratios based on the reported standard deviations.
• Benchmark predictive performance (R²/MAE) for models applied to four distinct locations.
• Analyze the trade-off between model performance metric and prediction error dispersion as described in the summary.

Strengths

• Performance metrics are reported with both mean and standard deviation, providing a measure of dispersion.
• The dataset directly compares results across four distinct locations.
• Data is provided in a standard XLS format with a size of 9.5 KB.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset's small size (9.5 KB) suggests a very limited scope, likely containing only summary statistics.

Provenance

Source

figshare

Collection Method

Likely contains aggregated results from machine learning experiments.

Freshness

Last updated 2026-04-30 17:54:55; freshness should be verified.

Geography

Covers four unspecified locations.