Performance metrics for machine learning models developed for quantitative structure–activity/property relationship (QSAR/QSPR) analysis of Gastroesophageal Reflux Disease (GERD) drug compounds. The dataset, authored by Mythili V and last updated on 2026-04-19, includes metrics such as R², RMSE, MAE, and Q² for models like SVR and Linear Regression. Each row corresponds to a model, detailing its target property, feature set, and cross-validation strategy.
Use Cases
- Benchmarking machine learning algorithms for QSAR/QSPR based on reported R², RMSE, and MAE metrics.
- Selecting feature sets or descriptor types for drug property prediction based on model performance comparisons.
- Evaluating the effectiveness of Leave-One-Out Cross-Validation (LOOCV) strategies in cheminformatics model validation.
- Analyzing the relationship between specific target properties of GERD drugs and model accuracy.
Strengths
- Includes multiple validation metrics per model: R², RMSE, MAE, and Q².
- Specifies the cross-validation strategy used (LOOCV).
- Released under a permissive CC-BY-4.0 license.
Limitations
- Row count is unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- figshare, author Mythili V.
- Collection Method
- Likely compiled from the results of QSAR/QSPR modeling experiments.
- Freshness
- Last updated 2026-04-19 14:09:50; freshness should be verified.