Understanding Breast Cancer Wisconsin (diagnostic) Dataset

The Breast Cancer Wisconsin (Diagnostic) dataset is a well-known dataset commonly used in machine learning. The dataset was curated by Dr. William H. Wolberg, W. Nick Street, and Olvi L. Mangasarian. It contains features computed from digitized images of fine needle aspirate (FNA) samples of breast mass tissue.

Characteristics of Breast Cancer Wisconsin (diagnostic) Dataset

1. Number of Instances: 569
2. Number of Attributes: 30 numerical attributes used for prediction, along with a class label.
3. Class Distribution: 212 – Malignant, 357 – Benign

Attributes of Breast Cancer Wisconsin (diagnostic) Dataset

The dataset comprises 30 features, including mean, standard error, and “worst” or largest values, computed for each image. These features encapsulate various aspects of cell nuclei characteristics:

1. mean radius: Mean of distances from center to points on the perimeter.
2. mean texture: Standard deviation of gray-scale values.
3. mean perimeter: Perimeter of the tumor.
4. mean area: Area of the tumor.
5. mean smoothness: Variation in radius lengths.
6. mean compactness: Perimeter^2 / Area – 1.0.
7. mean concavity: Severity of concave portions of the contour.
8. mean concave points: Number of concave portions of the contour.
9. mean symmetry: Symmetry of the cell nuclei.
10. mean fractal dimension: “Coastline approximation” – 1

The Breast Cancer Wisconsin (Diagnostic) dataset is a renowned collection of data used extensively in machine learning and medical research. Originating from digitized images of fine needle aspirates (FNA) of breast masses, this dataset facilitates the analysis of cell nuclei characteristics to aid in the diagnosis of breast cancer. In this article, we delve into the attributes, statistics, and significance of this dataset.