Cancer Visual Characteristics Dataset

Columns

• id: A unique identifier for each patient [1].
• diagnosis: Indicates the type of cancer, with 'M' for Malignant (Benign) and 'B' for Benign (Malignant) [1].
• radius_mean: The mean value of the cancer's radius [3, 4].
• texture_mean: The mean value of the cancer's texture [3, 5].
• perimeter_mean: The mean value of the cancer's perimeter [3, 5].
• area_mean: The mean value of the cancer's area [3, 6].
• smoothness_mean: The mean value of the cancer's smoothness [3, 7].
• compactness_mean: The mean value of the cancer's compactness [3, 7].
• concavity_mean: The mean value of the cancer's concavity [3, 8].
• concave points_mean: The mean value of the cancer's concave points [3, 9].
• symmetry_mean: The mean value of the cancer's symmetry [9].
• fractal_dimension_mean: The mean value of the cancer's fractal dimension [10].
• radius_se: The standard error of the cancer's radius [11].
• texture_se: The standard error of the cancer's texture [11].
• perimeter_se: The standard error of the cancer's perimeter [12].
• area_se: The standard error of the cancer's area [13].
• smoothness_se: The standard error of the cancer's smoothness [13].
• compactness_se: The standard error of the cancer's compactness [14].
• concavity_se: The standard error of the cancer's concavity [14].
• concave points_se: The standard error of the cancer's concave points [15].
• symmetry_se: The standard error of the cancer's symmetry [15].
• fractal_dimension_se: The standard error of the cancer's fractal dimension [16].
• radius_worst: The "worst" or largest mean value of the cancer's radius [16].
• texture_worst: The "worst" or largest mean value of the cancer's texture [17].
• perimeter_worst: The "worst" or largest mean value of the cancer's perimeter [18].
• area_worst: The "worst" or largest mean value of the cancer's area [18].
• smoothness_worst: The "worst" or largest mean value of the cancer's smoothness [19].
• compactness_worst: The "worst" or largest mean value of the cancer's compactness [20].
• concavity_worst: The "worst" or largest mean value of the cancer's concavity [20].
• concave points_worst: The "worst" or largest mean value of the cancer's concave points [21].
• symmetry_worst: The "worst" or largest mean value of the cancer's symmetry [21].
• fractal_dimension_worst: The "worst" or largest mean value of the cancer's fractal dimension [22].

Distribution

Usage

• Binary classification problems, such as predicting cancer type (Malignant or Benign) using visual features [24].
• Applying K-Nearest Neighbors (KNN) to diagnose cancer by considering neighbourhood relationships among patients with similar characteristics [24].
• Utilising Support Vector Machines (SVM) for classification tasks, particularly in two-class problems, by focusing on the clear separation of cancer types [25].
• Logistic Regression can be applied effectively for predicting cancer type [24].

Coverage

License

Who Can Use It

• Machine learning practitioners and researchers who aim to train or test algorithms for medical diagnosis [2].
• Data scientists looking to apply classification techniques like Logistic Regression, KNN, or SVM to real-world medical data [24, 25].
• Students and educators for educational purposes, exploring data analysis and machine learning in a healthcare context [26].
• Anyone interested in improving cancer-related visual features and diagnostic accuracy [2].

Dataset Name Suggestions

• Breast Cancer Diagnostic Features
• Breast Cancer Classification Data
• Cancer Visual Characteristics Dataset
• Medical Image Feature Analysis for Cancer
• Wisconsin Breast Cancer (Diagnostic) Data

Attributes