基于深度学习的新闻文本分类算法的研究与实现
首发时间:2017-12-12
摘要:文本分类属于自然语言处理领域一个非常典型的问题,应用普遍。使用深度学习算法处理文本分类任务可以避免繁杂的人工特征工程,但是如果单纯的使用卷积神经网络(CNN)或者长短时记忆网络(LSTM)模型,则会有长距离语义信息无法兼顾和训练不够高效的问题。本文提出将CNN与LSTM网络进行融合,借助两个模型各自的优势来提升文本分类模型的效果。本文经过多次实验对比分析,该模型比单一的神经网络模型的分类精度有明显提升,说明了该融合模型的有效性。
For information in English, please click here
Research and Implementation of News Text Classification Algorithm Based on Deep Learning
Abstract:Text categorization is a very typical problem in the field of natural language processing,which is commonly used.The use of deep learning algorithms to deal with text categorization tasks can avoid complicated manual feature engineering. However, if we simply use convolutional neural networks or longshort-term memory networks, it will be difficult to reconcile long-range semantic information or less efficient for training. This paper proposes to integrate CNN and LSTM networks, which is designed to improve the text classification model by combining the respective advantages of the two networks models. This article sets some comparative experimental. Compared with a single neural network model, the text classification model has obvious improvement in classification accuracy, which illustrate the validity of the model.
Keywords: Text Classification Deep Learning Convolutional Neural Networks Long Short-term Memory
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
基于深度学习的新闻文本分类算法的研究与实现
评论
全部评论0/1000