Breast Cancer Prediction using Scikit-Learn and Seaborn

1. Software used

• Python with ML Libraries installed
• VS Code Environmet

2. Required Dataset

Source - Kaggle

Attribute Information:

1. ID number
2. Diagnosis (M = malignant, B = benign)

• (3 – 32) Ten real-valued features are computed for each cell nucleus:

1. radius (mean of distances from center to points on the perimeter)
2. texture (standard deviation of gray-scale values)
3. perimeter
4. area
5. smoothness (local variation in radius lengths)
6. compactness (perimeter^2 / area - 1.0)
7. concavity (severity of concave portions of the contour)
8. concave points (number of concave portions of the contour)
9. symmetry
10. fractal dimension ("coastline approximation" - 1)

• The mean, standard error and "worst" or largest (mean of the three largest values) of these features were computed for each image, resulting in 30 features. For instance, field 3 is Mean Radius, field 13 is Radius SE, field 23 is Worst Radius.

• All feature values are recoded with four significant digits.

• Headers of the dataset:

3. EDA and Label Encoding

• Exploratory data analysis was performed using Pandas. Column with missing values was dropped.
• Categorical variable is converted to numerical values.

4. Cancer Prediction using Logistic Regression

• Dataset is split into training and test set. 75% of the data was used for training while remaining 25% was used for test.

• Logistic Regression package is imported from Scikit-Learn and applied to get prediction on the presence of cancer.

• The predicted values is plotted as a heatmap of the Confusion Matrix using Seaborn Library to determine the number of Type I and Type II errors.

• Confusion Matrix :

Calculated accuracy_score of the prediction: 97.89%