应用SD-LS-SVM算法的评论情感分析模型  

Comment Sentiment Analysis Model Using SD-LS-SVM Algorithm

作　　者：孙翌博 濮泽堃 徐玉华[3] 胡冰[2] SUN Yi-hao;PU Ze-kun;XU Yu-hua;HU Bing(School of Computer Science and Information Engineering,Changzhou Institute of Technology,Changzhou 213032,China;Post Big Data Technology and Application Engineering Research Center of Jiangsu Province,Nanjing University of Posts and Telecommunications;Post Industry Technology Research and Development Center of the State Posts Bureau(Internet of Things Technology),Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]常州工学院计算机信息工程学院,江苏常州213032 [2]南京邮电大学江苏省邮政大数据技术与应用工程研究中心 [3]南京邮电大学国家邮政局邮政行业技术研发中心(物联网技术),江苏南京210003

出　　处：《软件导刊》2021年第4期43-48,共6页Software Guide

基　　金：国家自然科学基金项目(61802200,61702281)。

摘　　要：随着电商经济的快速发展,消费者评论的情感预测有助于电商平台和商家研究销售策略实现精准营销。提出一种应用SD-LS-SVM算法的评论情感分析模型。首先构建词网,并利用上下文分析技术计算待检测评论中分词的评分,提取评论数据的特征向量。对LS-SVM进行基于置信区间简单动态优化的向量修剪,由改进的SD-LS-SVM算法对评论数据进行情感分类。为了验证模型的有效性,设计仿真实验对模型训练与预测结果进行统计分析。实验结果证明,该模型可以对评论文本进行有效情感分类且准确率达70%~85%。With the rapid development of the e-commerce economy,emotional prediction of consumer reviews helps e-commerce plat⁃forms and merchants formulate sales strategies to achieve precision marketing.This paper proposes a comment sentiment analysis mod⁃el using SD-LS-SVM algorithm.A word network is built to calculate the score of the word inthecomment with context analysis technolo⁃gy,and sentiment feature vector is extracted.Vector pruning is carried out based on simple optimization of dynamic confidence interval for LS-SVM(Least Squares Support Vector Machines).The improved SD-LS-SVM algorithm is used for sentiment classification of comment data.In order to verify the effectiveness of the model,this paper designs a simulation experiment to analyze the training and prediction results of the model.The experimental results prove that the model can effectively classify the sentiment of the review text.The experimental results prove that the model can effectively classify the sentiment of the review text with an accuracy rate of 70%to 85%.

关 键 词：情感分析 上下文分析 置信区间 最小二乘支持向量机 

分 类 号：TP306[自动化与计算机技术—计算机系统结构]