针对卷积神经网络聚焦于局部特征,不足以捕捉文本中长程依赖关系的问题,本文提出了一种基于CNN和自注意力机制改进的双通道图书标签分类模型(Gate Convolution Neural Network based on self- attention mechanism, GCNN-SAM)。该模型使...针对卷积神经网络聚焦于局部特征,不足以捕捉文本中长程依赖关系的问题,本文提出了一种基于CNN和自注意力机制改进的双通道图书标签分类模型(Gate Convolution Neural Network based on self- attention mechanism, GCNN-SAM)。该模型使用skip-gram将词嵌入成稠密低纬的向量,得到文本嵌入矩阵,分别输入到门卷积神经网络和自注意力机制,再经过逐点卷积,将两个通道中经过特征提取层得到的特征进行融合用于图书标签分类。在复旦大学中文文本分类数据集上进行对比实验,相较于SCNN、GCNN和其它改进的模型,测试集准确率达到96.21%,表明了GCNN-SAM模型在图书标签分类上具有优越性。同时,为验证GCNN-SAM模型的有效性,消融实验结果表明GCNN-SAM模型相较于CNN、GCNN和CNN-SAM在分类准确率上分别提升了5.9%、3.19%和3.66%。展开更多
Turnout is one of the important signal infrastructure equipment,which will directly affect the safety and efficiency of driving.Base on analysis of the power curve of the turnout,we extract and select the time domain ...Turnout is one of the important signal infrastructure equipment,which will directly affect the safety and efficiency of driving.Base on analysis of the power curve of the turnout,we extract and select the time domain and Haar wavelet transform characteristics of the curve firstly.Then the correlation between the degradation state and the fault state is established by using the clustering algorithm and the Pearson correlation coefficient.Finally,the convolutional neural network(CNN)and the gated recurrent unit(GRU)are used to establish the state prediction model of the turnout to realize the failure prediction.The CNN can directly extract features from the original data of the turnout and reduce the dimension,which simplifies the prediction process.Due to its unique gate structure and time series processing features,GRU has certain advantages over the traditional forecasting methods in terms of prediction accuracy and time.The experimental results show that the accuracy of prediction can reach 94.2%when the feature matrix adopts 40-dimensional input and iterates 50 times.展开更多
文摘针对卷积神经网络聚焦于局部特征,不足以捕捉文本中长程依赖关系的问题,本文提出了一种基于CNN和自注意力机制改进的双通道图书标签分类模型(Gate Convolution Neural Network based on self- attention mechanism, GCNN-SAM)。该模型使用skip-gram将词嵌入成稠密低纬的向量,得到文本嵌入矩阵,分别输入到门卷积神经网络和自注意力机制,再经过逐点卷积,将两个通道中经过特征提取层得到的特征进行融合用于图书标签分类。在复旦大学中文文本分类数据集上进行对比实验,相较于SCNN、GCNN和其它改进的模型,测试集准确率达到96.21%,表明了GCNN-SAM模型在图书标签分类上具有优越性。同时,为验证GCNN-SAM模型的有效性,消融实验结果表明GCNN-SAM模型相较于CNN、GCNN和CNN-SAM在分类准确率上分别提升了5.9%、3.19%和3.66%。
基金National Natural Science Foundation of China(Nos.61863024,71761023)Funding for Scientific Research Projects of Colleges and Universities in Gansu Province(Nos.2018C-11,2018A-22)Natural Science Fund of Gansu Province(No.18JR3RA130)。
文摘Turnout is one of the important signal infrastructure equipment,which will directly affect the safety and efficiency of driving.Base on analysis of the power curve of the turnout,we extract and select the time domain and Haar wavelet transform characteristics of the curve firstly.Then the correlation between the degradation state and the fault state is established by using the clustering algorithm and the Pearson correlation coefficient.Finally,the convolutional neural network(CNN)and the gated recurrent unit(GRU)are used to establish the state prediction model of the turnout to realize the failure prediction.The CNN can directly extract features from the original data of the turnout and reduce the dimension,which simplifies the prediction process.Due to its unique gate structure and time series processing features,GRU has certain advantages over the traditional forecasting methods in terms of prediction accuracy and time.The experimental results show that the accuracy of prediction can reach 94.2%when the feature matrix adopts 40-dimensional input and iterates 50 times.
