The project was carried out on the breast cancer dataset compiled for research. It can be found at: UCI ML Repository and also on Kaggle
- ID number
- Diagnosis (M = malignant, B = benign) 3-32)
Ten real-valued features are computed for each cell nucleus:
a) radius (mean of distances from center to points on the perimeter) b) texture (standard deviation of gray-scale values) c) perimeter d) area e) smoothness (local variation in radius lengths) f) compactness (perimeter^2 / area - 1.0) g) concavity (severity of concave portions of the contour) h) concave points (number of concave portions of the contour) i) symmetry j) 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. | Missing attribute values: none | Class distribution: 357 benign, 212 malignant
A SVM Classifier was used. Preprocessing and EDA was carried out and the 26 best parameters that affected the prediction were chosen. A little bit of tweaking on the C parameter and use of rbf kernel yielded better results as compared to a linear kernel. The scores obtained were as follows:
- Accuracy = 0.93
- Precision = 0.95
- Recall = 0.74
- F1-Score = 0.83
The score can be improved on further analysis and experimentation with various kernels and tweaking of 'C' and 'gamma' parameters.