基于集成学习的神经网络对小RNA的预测

摘要

The way of computing intelligence,that is,particle swarm optimized feedforward artificial neural network and their integrated approach,has been applied in the forecasting of microRNA. We use 36 features extracted from the pre-microRNA sequences to train the neural network model,and test the real and pseudo data sets using the trained model with over 90.0% average prediction accuracy. Furthermore,higher accuracy has been got by the ensemble learning. The result indicates that the approach of neural network to predict microRNA is an effective method.%通过计算智能的方法,即粒子群算法优化的前向人工神经网络和集成神经网络来预测小RNA.从小RNA前体序列中提取36维特征用于训练神经网络模型,利用训练好的模型分别对阳性和阴性小RNA数据集进行测试,预测平均精度达到90％以上,并且通过集成神经网络的方法取得了更高的精度.实验结果表明,利用集成神经网络建模进行小RNA的预测是一条行之有效的途径.