计算机与现代化

• 人工智能 • 上一篇    下一篇

基于VAE-DBN双模型的智能文本分类方法

  

  1. (1.军事科学院研究生部,北京100091;2.31608部队,福建厦门361025)
  • 收稿日期:2018-05-21 出版日期:2019-01-03 发布日期:2019-01-04
  • 作者简介:王玮(1975-),男,内蒙古清水河人,军事科学院研究生部博士研究生,31608部队工程师,研究方向:智能数据工程。

Intelligent Text Classification Method Based on VAE-DBN Dual-Model

  1. (Dept. of Graduate, Academy of Military Sciences, Beijing 100091, China; 2. 31608 Force, Xiamen 361025, China)
  • Received:2018-05-21 Online:2019-01-03 Published:2019-01-04

摘要: 文本分类技术是信息过滤、搜索引擎等领域的基础,是当下研究热点之一。本文在介绍文本分类相关概念、深度学习相关模型的基础上,通过分析传统文本分类方法存在的不足,提出基于变分自编码器模型和深度置信网络模型(VAE-DBN)的双模型融合的文本分类方法。通过在相关语料集上的对比验证,表明该双模型方法能有效提高文本分类的准确性。

关键词: 变分自编码器, 深度置信网络, 文本分类

Abstract: Text categorization technology is the foundation of information filtering, search engine and other fields, and is one of current research hot-spots. Based on the introduction of text classification related concepts and deep learning related models, this paper presents a dual-model text classification method based on the variational autoencoder model and the deep belief network model (VAE-DBN) by analyzing the shortcomings of the traditional text classification methods. By comparing and verifying the corpus, the results show that the dual-model method can effectively improve the accuracy of text categorization.

Key words: variational autoencoder, deep belief network, text categorization

中图分类号: