Wisconsin Breast Cancer Cytology Features for Benign/Malignant Classification

Cytology features from breast cancer biopsies, including clump thickness, uniformity of cell size and shape, marginal adhesion, and other cellular characteristics. The data was obtained from the UCI Machine Learning Repository and is used to predict whether a biopsy is benign or malignant. Each of the nine predictive features is scored on a scale from 1 to 10.

Use Cases

• Train a binary classifier to predict breast cancer (benign vs. malignant) based on nine cytology feature scores.
• Benchmark feature selection or model performance using a well-known medical dataset with ordinal feature scales.
• Analyze the relationship between specific cellular characteristics, like bare nuclei or mitoses, and cancer diagnosis.

Strengths

• Features are clearly defined with ordinal scales (1-10) for nine cytological attributes.
• Includes a straightforward binary class label (0 for benign, 1 for malignant) for supervised learning.
• Data provenance is documented, citing the original UCI Machine Learning Repository source.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• Column-level documentation is absent in this metadata; field semantics must be fully verified after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

UCI Machine Learning Repository (University of California, Irvine).

Collection Method

Features extracted from breast cancer biopsy cytology.