摘要
新发展格局下,实体企业依托电商平台零售商品已成为我国电子商务的主要模式。良好的网络产品交流平台对企业提升电子商务竞争力具有积极作用。选取包含评论者和评论内容的多类型评论特征指标,构建基于ID3算法的网络产品交易非理性评论识别模型,并对具体案例中的非理性评论进行了识别。研究表明,网络产品交易非理性评论识别模型的总体识别效果较好,F 1值达到73%,没有出现过拟合现象,泛化能力较好。本方法优于随机森林、BP神经网络和支持向量机等机器学习方法,能够较好地解决以离散型特征指标为主的非理性评论的识别问题。
Under the new development pattern,entity enterprises relying on e-commerce platform to retail goods has become the main mode of e-commerce in China.A good commodity use exchange platform has a positive effect on enterprises to enhance the competitiveness of e-commerce.In the paper,we select multi-type review feature indicators containing reviewers and review contents,construct an irrational review identification model for online product transactions by ID3 algorithm,and identify irrational reviews in specific cases.The study shows that the overall recognition effect of the irrational comment recognition model for online product transactions is good,with the F1 value reaching 73%,no overfitting phenomenon and good generalization ability.This method outperforms machine learning methods such as random forest,BP neural network and support vector machine,and can better solve the recognition problem of irrational comments with discrete feature indicators.
出处
《阅江学刊》
2023年第4期105-117,173,共14页
Yuejiang Academic Journal
基金
中国社会科学院大学(研究生院)研究生科研创新支持计划项目“碳市场对低碳技术创新效用的测度研究”(2023-KY-55)。
