Research and application of pca-bp-bagging model in medical assistant diagnosis

In this paper, an algorithm that based on pca-bp-bagging model is developed for the prediction of pathological data. This algorithm aims at improving the characteristics of bp neural network that the prediction accuracy of pathological data is low, the generalization ability of single bp neural network model is poor, and the anti-interference ability is weak. To enhance the performance of the whole model, it combines the idea of neural network ensemble learning with bp neural network. By analyzing the pathologic data of breast cancer patients from the University of California Irvine, it achieves the ideal forecast results. The experimental results show that the algorithm is effective, in addition, the pca-bp-bagging has higher prediction accuracy than other classical methods.