Analyzing the discriminative attributes of products using text mining focused on cosmetic reviews
作者：
Highlights：
• Text mining for online reviews analyzed customer preferences for detailed attributes of products.
• We calculated the percentage of words in a review and extracted the discriminative attributes of competitors' products.
• We analyzed the attitude and preference of the customers about the attributes of the product by calculating the probability of word appearing in the rating.
• We propose an efficient market research method by using ratio analysis of words, latent semantic analysis, and LDA sequentially.
摘要
•Text mining for online reviews analyzed customer preferences for detailed attributes of products.•We calculated the percentage of words in a review and extracted the discriminative attributes of competitors' products.•We analyzed the attitude and preference of the customers about the attributes of the product by calculating the probability of word appearing in the rating.•We propose an efficient market research method by using ratio analysis of words, latent semantic analysis, and LDA sequentially.
论文关键词：Text mining,Latent semantic analysis,LSA,Labeled latent Dirichlet allocation,L-LDA,Discriminative attributes of products
论文评审过程：Received 6 February 2018, Revised 18 May 2018, Accepted 1 June 2018, Available online 2 July 2018, Version of Record 2 July 2018.
论文官网地址：https://doi.org/10.1016/j.ipm.2018.06.003
