Airlines: Subsampled Flight Data for Classification with 10 Classes

A subsampled version of the OpenML airlines dataset, generated with a specific random seed. The subset likely contains up to 2000 rows and 100 columns, with the target variable limited to 10 classes. It was created by Eddie Bergman and is shared under a US public domain license.

Use Cases

• Benchmarking classification model performance based on the described 10-class target variable.
• Feature selection experiments based on the described subset of 100 columns.
• Testing data sampling and stratification techniques based on the generation method described.

Strengths

• Subsampling parameters are explicitly defined: seed=4, nrows_max=2000, ncols_max=100, nclasses_max=10.
• Generation code is provided, allowing for reproducibility of the sampling process.
• Derived from a known OpenML dataset (ID 1169), providing a traceable origin.

Limitations

• Row count is unknown, which may limit suitability assessment for specific modeling needs.
• 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

OpenML (dataset ID 1169)

Collection Method

Algorithmic subsampling of the original 'airlines' dataset using provided Python code.