An Optimized Random Forest Classification Method for Processing Imbalanced Data Sets of Alzheimer's Disease

In this work we propose an optimized random forest classification method to solve the problem of imbalanced samples in the Alzheimer's disease (AD). The improved algorithm is based on Synthetic Minority Oversampling Technique (SMOTE) and multi-dimensional Gaussian probability density hypothesis combined with Random Forest (RF). Through this technique, some reasonable pseudo samples can be established for minority classes to balance the number of samples of each type before the RF classifier is used to classify them. This method can effectively solve the problem of sample imbalance, improve the accuracy and recall rate of the minority classes to a certain extent, and avoid overfitting of the classifier. For the purpose of demonstrating the effectiveness of the algorithm proposed in this paper, Support Vector Machine (SVM) algorithm, RF algorithm and the algorithm proposed in this paper are compared. Experimental results have proved that the algorithm proposed in this paper has obvious improvement in accuracy and recall rate compared with the other two algorithms.