基于SVR支持向量回归模型的在线商品评论信息可信度分析研究
关键词 商品评论信息 可信度量化 SVR 机器学习 数据挖掘
毕业论文外文摘要
Title Analysis of Online Product Reviews Credibility Based on
Support Vector Regression model
Abstract
The popularity of online shopping makes product reviews received rising attention, researchers about reviews information credibility has become a hotspot of the current text field. Through in-depth analysis of current research on reviews information credibility , understanding the existing problems in the Research on reviews information credibility measurement and then selected the support vector regression algorithm whose output value can be any real number to quantify the reviews information credibility. Firstly, according to the empirical research of reviews information credibility factors , they selected content integrity, sentimental balance, review timeliness and credit rating of the publisher as four features dimensions that can be quantified. Secondly, build the feature dictionary and develop the credibility characteristic quantification processing software to complete feature quantification ; Thirdly, with the help of machine learning methods, they use LIBSVM tools to complete reviews credibility calculation. Finally, use the Grid Search method to revise the parameters , in order to get the optimize results. After matching analysis the experimental results with manual annotation average and reviews content, it was found that the credibility measurement model forecasts the credibility value and the type of product reviews effectively.
Keywords product reviews credibility measurement SVR machine learning
data mining
目 次
1 绪论 1
1.1 研究背景及意义 1
1.1.1 研究背景 1
1.1.2 研究意义 2
1.2 研究现状 3
1.2.1 商品评论信息的可信度影响因素研究现状 4
1.2.2 商品评论信息可信度计量研究现状 5
1.2.3 现有研究存在的问题 7
1.3 本文的主要研究内容 7
1.4 本研究的特色 8
下一篇：正则半群的性质及应用+文献综述