摘 要:在互联网高度发展和智能技术普及的大环境下,电商平台出现了大量的评论数据,它们对挖掘用户需求和建立商品口碑具有重要价值。文章爬取了京东电商平台上某品牌手机的评论数据,并基于预处理之后的数据进行了倾向性分析和 LDA主题模型分析。研究结果表明,该品牌手机具有外观好看、充电快、性价比高和拍照功能强大等优势,但也有新品定价偏贵、保值率低、售后服务差、部分包装零件不全等不足之处。所得结论为该品牌手机升级提供一定的参考依据。
关键词:倾向性分析;LDA 主题模型;品牌手机
DOI:10.19850/j.cnki.2096-4706.2023.02.003
基金项目:广西大学生创新创业项目(201910595202)
中图分类号:TP181 文献标识码:A 文章编号:2096-4706(2023)02-0012-03
Analysis of Comment Data of a Brand Mobile Phone Based on LDA Theme Model
WU Nannan, SHI Jiacheng, LIU Shengqiang
(School of Mathematical & Computing Science, Guilin University of Electronic Technology, Guilin 541004, China)
Abstract: In the context of the high development of the Internet and the popularization of intelligent technology, a large number of review data have emerged on E-commerce platforms, which are of great value in mining user needs and establishing product reputation. It crawls the review data of a brand's mobile phone on JD E-commerce platform, and conducts a tendentiousness analysis and LDA theme model analysis based on the pre processed data. The research results show that the mobile phone of this brand has the advantages of goodlooking appearance, fast charging, high cost performance and strong photographing function, but it also has the disadvantages of expensive new product pricing, low value preservation rate, poor after-sales service, and incomplete parts of some packaging. The conclusion provides a reference for the upgrading of the mobile phone of this brand.
Keywords: tendentiousness analysis; LDA theme model; brand mobile phone
参考文献:
[1] 刘玉林,菅利荣 . 基于文本情感分析的电商在线评论数据挖掘 [J]. 统计与信息论坛,2018,33(12):119-124.
[2] 李天辰,殷建平 . 基于主题聚类的情感极性判别方法 [J].计算机科学与探索,2016,10(7):989-994.
[3] BLEI D M,NG A Y,JORDAN M I. Latent Dirichlet Allocation [J].Journal of Machine Learning Research,2003(3):993-1022.
[4] 王鹏飞,张斌 . 基于文献计量的国内 LDA 主题模型研究进展分析 [J]. 图文情报研究,2020,13(2):85-91+111.
[5] 王丽君 . 词向量和文本隐含主题的联合学习研究 [D]. 武汉:华中师范大学,2018.
[6] 袁扬,李晓,杨雅婷 . 基于 LDA 主题模型的维吾尔语无监督词义消岐 [J]. 厦门大学学报:自然科学报,2020,59(2),198-205.
作者简介:吴楠楠(2002.05—),男,汉族,湖北武穴人,本科在读,研究方向:数据分析;石家程(2001.11—),男,汉族,海南乐东人,本科在读,研究方向:数据分析;刘胜强(1998.01—),男,汉族,广西桂林人,JAVA 技术顾问,本科,研究方向:软件开发。