论文部分内容阅读
角分类算法是一类快速分类算法,以其为学习算法的前向神经网络,在信息检索,特别是在线信息检索等领域有着重要的应用.通过对CC4学习算法的分析,揭示了泛化距离在角分类神经网络中的意义.针对文本数据的快速分类要求,提出了新的角分类网络TextCC.为解决数据的多类别判定问题,给出了新的角分类神经网络隐层与输出层之间连接矩阵的学习算法.实验表明,新的角分类神经网络隐层与输出层之间连接矩阵的学习算法有效,TextCC的分类精度教CC4的分类精度显著的提高.
Angle classification algorithm is a kind of fast classification algorithm, which is an advanced neural network learning algorithm, and has an important application in information retrieval, especially online information retrieval and other fields.By analyzing the CC4 learning algorithm, the generalization distance In order to solve the problem of multi-category decision of data, a new angular classification neural network hidden layer and output layer is proposed The experimental results show that the learning algorithm of the connection matrix between the hidden layer and the output layer of the new angular classification neural network is effective, and the classification accuracy of TextCC classifier CC4 is significantly improved.