论文部分内容阅读
有效地产生泛化能力强、差异大的个体学习器,是集成学习算法的关键.为了提高学习器的差异性和精度,文中提出一种基于成对差异性度量的选择性集成方法.同时研究一种改进方法,进一步提高方法的运算速度,且支持并行计算.最后通过使用BP神经网络作为基学习器,在UCI数据集上进行实验,并与Bagging、基于遗传算法的选择性集成(GASEN)算法进行比较.实验结果表明,该改进算法在性能上与GASEN算法相近的前提下,训练速度得到大幅提高.
To effectively produce individual learners with strong generalization ability and large differences is the key to the integrated learning algorithm.In order to improve the learner’s difference and accuracy, a selective integration method based on pairwise difference measure is proposed in this paper.At the same time, An improved method to further improve the computing speed of the method and support parallel computing.Finally, experiments were performed on the UCI dataset by using BP neural network as a learner, and compared with Bagging, GASEN (Genetic Algorithm for Selective Integration) The experimental results show that the performance of the improved algorithm is greatly improved compared with GASEN algorithm.