【24h】

Fake News Detection Method Based on Text-Features

机译:基于文本特征的假新闻检测方法

获取原文
获取外文期刊封面目录资料

摘要

Feature extraction is a critical task in fake news detection. Embedding techniques, such as word embedding and deep neural networks, are attracting much attention for textual feature extraction, and have the potential to learn better repre-sentations. In this paper, we propose a joint Convolutional Neural Network model (CNN) and a Long Short Term Memory (LSTM) recurrent neural network architecture, taking advantage of the coarse-grained local features generated by CNN and long-distance dependencies learned via LSTM. An empirical evaluation of our model shows good prediction accuracy of fake news detection, when compared to Support Vector Machine and CNN baselines.
机译:特征提取是假新闻检测中的关键任务。嵌入技术,如Word嵌入和深神经网络,是对文本特征提取的巨大关注,并且有可能学习更好的代价。在本文中,我们提出了一个关节卷积神经网络模型(CNN)和长期内存(LSTM)复发性神经网络架构,利用CNN和通过LSTM学习的长距离依赖性产生的粗粒粒度局部特征。与支持向量机和CNN基线相比,我们模型的实证评估显示了假新闻检测的良好预测准确性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号