基于词向量和CNN的书籍评论情感分析方法  

Sentiment Analysis Method of Book Reviews Based on Word Vector and Convolution Neural Network

作　　者：李昊璇[1] 张华洁 LI Haoxuan;ZHANG Huajie(College of Physics and Electronic Engineering, Shanxi University, Taiyuan 030006, China)

机构地区：[1]山西大学物理电子工程学院,山西太原030006

出　　处：《测试技术学报》2019年第2期165-171,共7页Journal of Test and Measurement Technology

摘　　要：针对在线书籍评论的情感分析问题,基于词向量和深度学习原理,提出了一种基于词向量和卷积神经网络(Convolutional Neural Networks, CNN)的书籍评论情感分类方法.实验结果表明,本方法对在线书籍评论的情感分类准确率达到92.99%,同时,此方法对于大量文本集的情感分析有很好的适用性.此外还研究了不同语料库、书籍评论词向量的维度、书籍评论数据集的大小等对情感分析结果的影响,实验结果显示,构建针对性的语料库,基于词向量和卷积神经网络的书籍评论情感分析方法是一种简单而有效的文本情感分析方法,该方法具有扩展性和对不同评论文本的适用性.Aiming at the emotional analysis of online book reviews, an emotional classification method of book reviews is proposed based on word vector and deep learning principle in this paper. The experimental results show that the method achieves an accuracy of 92.99% for online book reviews, and it is applicable to large text sets. In addition, this paper also studies the influence of the dimensions of different corpus and book commentary vectors the proportion of emotion in the data set and the size of book review data set on the result of sentiment analysis. The experimental results show that the method based on word vector and convolutional neural network is a simple and effective method for sentiment analysis of book reviews. The method has scalability and applicability to different review texts.

关 键 词：词向量 卷积神经网络 情感分析 语料库 数据挖掘 自然语言处理 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]