基于改进的卷积神经网络邮件分类算法研究

针对传统文本分类方法中出现的维度过高和数据稀疏问题,通过对卷积神经网络(Convolutional Neural Network,CNN)和 inception V1 模型的深入研究,将两个模型融合起来,提出了一种基于 i-CNN 模型的 邮件分类方法;在卷积、池化操作中加入了 1×1 卷积核降低特征向量的厚度,减少了参数,提高了计算性能; 通过数据验证,i-CNN 模型对邮件的分类结果高达 92. 18%,在对比实验中,i-CNN 模型相对于几种机器学 习分类模型,取得了最高的分类精准率,在有无 inception 结构模型对比中,i-CNN 模型精准率高于 CNN 模 型;说明该模型具有较好的分类效果,且 inception V1 模型的融入能提高文本分类的准确率。

Research on Mail Classification Algorithm Based on Improved Convolutional Neural Network

Aiming at the problems of high dimension and sparse data in traditional text classification methods, this paper proposes an e-mail classification method based on i-CNN model by combining convolutional neural network (CNN) and inception V1 model. In the convolution and pooling operation, 1 × 1 convolution kernel is added to reduce the thickness of eigenvectors, reduce the parameters and improve the computational performance. Through data validation, the result of i-CNN model for e-mail classification reaches as high as 92. 18%. In the comparative experiment, compared with several machine learning classification models, i-CNN model achieved the highest classification accuracy. In the comparison with or without the inception structure model, i-CNN model accuracy is higher than CNN model. It shows that the model has a good classification effect, and the integration of inception V1 model can improve the accuracy of text classification.