Aim: The objective of the work is to predict the Accuracy of Breast Cancer Prediction through Novel Convolutional Neural networks (NCNN) Comparative with the Relief Algorithm (RA). Material And Methods: The dataset used for Accuracy and Loss is from the GitHub library. The total sample size is 48. The two groups Novel Convolutional Neural Network (N=24), and Relief Algorithm (N=24) proposed by predicting the accuracy (97.10%) of Breast Cancer Prediction compared with Novel Convolutional Neural Network. Results: The Result proved that the Relief algorithm with Better accuracy than the Novel Convolutional Neural Network. The Novel Convolutional Neural Network appears significantly better than Relief Algorithm (p<0.05). Discussion and Conclusion: The Prediction of breast cancer is better in novel convolutional neural networks when compared with the relief algorithm (RA).