基于在线评论挖掘的宠物家具产品设计改进方法

A Design Improvement Method for Pet Furniture Products Based on Online Comment Mining

摘要: 为了获取用户对产品的满意度并用于产品的改进设计，提出了一种基于在线评论的设计改进方法。本文以宠物家具为研究对象，采用爬虫技术从电商平台获取了大量的在线评论；通过分词技术和词向量技术对在线评论进行了预处理；接着构建用户极性词典并借助狄利克雷主题模型确定了产品特征；最后分析挖掘了产品各个特征需改进程度从而建立了宠物家具评价指标体系，并找到亟需改进的产品特征，从而制定产品的改进策略。该方法将数据挖掘与情感分析结合，量化了产品需改进的程度，为设计师提供了有针对性的家具产品改进方案。

Abstract: In order to obtain user satisfaction with the product and use it for product improvement design, a design improvement method based on online review is proposed. Taking pet furniture as the research object and using crawler technology to obtain a large number of online comments from e-commerce platforms, the online comments are preprocessed by word segmentation technology and word vector technology; the user polarity dictionary is constructed and the product characteristics are determined with the help of Dirichlet theme model; the user satisfaction and the degree of improvement of each product feature are analyzed and excavated to establish the pet furniture evaluation index system, and find the product features that need to be improved, so as to formulate the product improvement strategy. This method combines data mining with emotion analysis, quantifies user satisfaction, and provides targeted product improvement direction for designers.