Performance Outcomes for PCA, t-SNE, and Random-Forest Feature Selection

Performance outcomes from applying dimensionality reduction and feature selection techniques to a microbial dataset. The dataset, created by Tong Gao, is available under a CC-BY-4.0 license and was last updated on March 19, 2026. It is a small file, 5.5 KB in size, stored in XLS format.

Use Cases

• Compare the performance of PCA and t-SNE for visualizing or preprocessing high-dimensional microbial data.
• Evaluate the effectiveness of random-forest feature selection for identifying key predictive features.
• Benchmark model robustness and predictive performance in the context of oil contamination prediction.
• Stress-test machine learning models using a limited set of top features derived from feature selection.

Strengths

• Dataset is openly licensed under CC-BY-4.0, permitting broad reuse.
• File size is small (5.5 KB), facilitating quick download and inspection.

Limitations

• Row count and specific column definitions are unknown, limiting suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare, author Tong Gao.

Collection Method

Likely contains results from computational experiments applying PCA, t-SNE, and random-forest feature selection.

Freshness

Last updated 2026-03-19 17:38:41; freshness should be verified.