摘要
使用情感分析算法获取在线评论的效价,并应用前景理论和熵值法对在线评论的效价进行调整,结合评论数量和网络搜索数据,建立了自回归分布滞后模型,并对不同价位档次汽车销量进行预测分析.研究发现,考虑了消费者受负面信息影响的在线评论的效价、数量和网络搜索数据的模型预测效果优于传统模型,更符合实际情况,但对不同价位档次汽车预测效果存在差异,低档汽车销量预测效果最佳,其次是中档汽车,最后为高档汽车.
The emotion analysis algorithm is used to obtain the valence of online reviews,and the prospect theory and entropy method are used to adjust the valence of online reviews.Combined with the volume of reviews and Internet search data,an autoregressive distributed lag model is established to predict and analyze the sales volume of cars in different price levels.It is shown that the prediction effect of the model,which takes into account the valence,volume of online reviews and Internet search data affected by negative information,is better than that of the traditional model and more consistent with actual situations.However,there are differences in the prediction effect for cars of different price levels.The sales prediction of low-end cars is the best,followed by mid-range cars,and finally high-end cars.
作者
王书田
林岩
朱国庆
闫叶金
WANG Shu-tian;LIN Yan;ZHU Guo-qing;YAN Ye-jin(School of Maritime Economics and Management,Dalian Maritime University,Dalian 116026,China)
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2023年第5期752-760,共9页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(72271037)。
关键词
在线评论
情感分析
网络搜索
销量预测
自回归分布滞后模型
online reviews
sentiment analysis
Internet search
sales forecasting
autoregressive distribution lag model
作者简介
王书田(1993-),女,山东聊城人,大连海事大学博士研究生;林岩(1972-),男,山东济宁人,大连海事大学教授,博士生导师.
