Aim: The objective of the study is to detect Breast Cancer with the help of Chest Diagnostic image dataset by using Machine Learning based on Naive Bayes Algorithm. To achieve accuracy, a novel augmented dataset classification is used. Materials and Methods: Accuracy and Loss of Breast cancer detection are performed using the kaggel library. Dataset of 220 chest image dataset with total sample size 20 is used and the two groups are Naive Bayes (N=10) and Novel Multilayer Perceptron (N=10). Results: This study proved that Naive Bayes achieved better accuracy of 99.66% which is higher, compared to Novel Multilayer Perceptron accuracy of 99.13%. Finally, the Naive Bayes appears significantly better than the Novel Multilayer Perceptron. The statistical analysis shows insignificant differences between the sample groups with p=0.184 (p<0.05) and confidence level of 95%. Conclusion: This study shows that Naive Bayes achieves better accuracy than the Novel Multilayer Perceptron.