Comparison Based Study to Predict Breast Cancer :A Survey

Note: Java implementation using Dl4j is added in a seperate folder.

Refer Grover, A., Pradhan, N., & Hemrajani, P. (2020). Comparison-Based Study to Predict Breast Cancer: A Survey for full details.

The following dataset has been taken from UCI Machine Learning Repository.:

Performing Exploratory Analysis

Pearson's Correlation of features with respect to the target variables represented here.

Neural Network Architechture

Adam optimizer has been used here.

• Input Layer : ReLU

• Hidden Layer: LeakyReLU

• Output Layer : Sigmoid

Results

Model	Accuracy	F1 Measure
CART	93.80	95.01
LR	94.76	95.87
SVM	94.76	95.52
HVC	95.23	96.23
XGBoost	96.67	97.37
ANN	95.23	96.76