Analyzing the Tendency of Review Expressions for the Automatic Construction of Cosmetic Evaluation Dictionaries
Analyzing the Tendency of Review Expressions for the Automatic Construction of Cosmetic Evaluation Dictionaries
复制标题
化妆品评价词典自动构建的评论表达倾向分析
DOI：
10.1142/9789811215094_0006
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Nakajima Shinsuke
中科院分区：
文献类型：
--
作者：
Ueda Mayumi;Taniguchi Yuna;Matsunami Yuki;Okuda Asami;Nakajima Shinsuke
顾客评论在网上购物体验中起着重要的作用，特别是对于化妆品等产品，其中产品的价值往往是主观的。然而，消费者要从大量的评论中筛选出对他们的购买决策有意义的评论是非常耗时的。推荐系统可以是一个有用的工具，帮助推荐产品给客户，这是相关的，他们的特定需求和背景。然而，化妆品产品有许多不同的属性（例如抗衰老效果和美白效果），这些属性不能用单个用户评分来表示。因此，在我们的研究中，我们的目标是开发一个推荐系统，考虑到这些属性时，向消费者推荐化妆品。本文介绍了如何化妆品评价字典可以构建基于审查表达式提取基于文本的评论意见，为不同的化妆品类别。
Customer Reviews play an important role in the online shopping experience, particularly for products such as Cosmetics where the value of the product tends to be subjective. However, it is time consuming for consumers to filter through the large number of reviews and identify those which are meaningful to their purchase decisions. Recommender systems could be a useful tool in helping recommend products to customers which are relevant to their particular needs and context. However, there are many different attributes of the cosmetic product (such as anti-ageing effect and whitening effect) which could not be represented with a single user score. As such, In our research, we aim to develop a recommender system which takes into account such attributes when recommending cosmetic products to consumers. This paper describes how a cosmetic evaluation dictionary could be constructed based on review expressions extracted from text-based review comments for different cosmetic categories.