论文部分内容阅读
【目的/意义】基于离散情感理论,对电商平台在线评论中所含不同离散情感的分布规律进行探究,发掘其对于营销管理的实践意义。【方法/过程】以手机这一搜索型产品的海量中文评论为研究对象,以情感认知模型OCC模型为情感分类依据,通过深度学习的方法构建离散情感语料库,并在此基础上对不同评论星级、不同的商品购买和评论发布的时间间隔中,评论所包含离散情感的分布特征进行了深入的研究。【结果/结论】研究发现:包含不同离散情感的评论在不同评论星级中的分布情况差别较大,在不同时间间隔中的分布曲线却大致相同,虽都与“长尾分布”非常类似,但仍有细微差别。
[Purpose / Significance] Based on the theory of discrete emotions, this paper explores the distribution of different discrete emotions contained in the e-commerce platform online reviews and explores its practical significance for marketing management. 【Method / Process】 Taking the massive Chinese commentary of the mobile phone as a search product as the research object and the emotion cognition model OCC model as the emotion classification basis, the dissertation constructs the discrete emotion corpus based on the deep learning method, and on this basis, Star, different product purchase and comment release time interval, the dissertation contains the characteristics of the distribution of discrete emotions conducted in-depth study. [Results / Conclusions] The study found that the distributions of commentary with different discrete emotions differ greatly among different commentary stars, and the distribution curves in different time intervals are roughly the same. Although they are all similar to the “long tail distribution” Similar, but there are still subtle differences.