Boruta is a wrapper algorithm for all relevant feature selection, developed by Miron B. Kursa. It identifies relevant features by comparing the importance of original attributes with the importance achievable at random, estimated using permuted copies called shadows. The dataset likely contains the algorithm's implementation details, performance benchmarks, or related research data.
Use Cases
- Benchmarking feature selection algorithms based on the described importance comparison method.
- Developing robust feature selection models based on the shadow attribute concept.
- Researching wrapper-based feature selection techniques for pattern recognition.
Strengths
- Algorithm developed by a named author, Miron B. Kursa.
- Description provides a clear conceptual overview of the feature selection method.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Last update date is unknown; freshness unverified.
Provenance
- Source
- Miron B. Kursa
- Collection Method
- null
- Time Range
- null
- Freshness
- null
- Geography
- null