社会促进者评分（SPS）和评论网络：一种预测在线购物品牌财务健康的方法和工具
Social Promoter Score (SPS) and Review Network: A Method and a Tool for Predicting Financial Health of an Online Shopping Brand
摘要：
总结顾客对在线购物品牌的产品的评价或情绪的传统方式不足以评估该品牌的财务健康状况。它忽视了个人客户的社会地位和影响力。在本文中，我们提出了一个名为评论网络的工具，用于测量客户在亚马逊等在线商品网站的影响力。使用这种测量的影响力，我们提出了一种方法，根据他们的评级和情绪，他们从在线商品网站上收集的评论来评估品牌的客户忠诚度。一个品牌的评论网络是由该品牌的所有产品的所有评论构建的，其中节点是客户，如果客户成为另一个客户撰写的评论的潜在读者，则会创建边缘。顾客在评论网络中的中心地位代表了她的影响力。我们提出的方法命名为社会促进者得分结合忠诚度和中心的所有客户的品牌。我们比较了我们的方法与基线方法的基础上的概念，净启动分数。我们在Amazon.com上对一些知名品牌的评论数据集进行了社会化推广评分。结果表明，社会促进者得分预测财务健康的品牌在未来的销售远远优于基线方法。我们已经注意到，一般情况下，社交推广得分的影响会在一到五个月内反映在产品销售上。
英文摘要：
The conventional way of summarizing ratings or sentiment of reviews of customers on products of an online shopping brand are not sufficient to evaluate the financial health of that brand. It overlooks the social standing and influence of individual customers. In this paper, we have proposed a tool named as Review Network for measuring the influence of customers in online merchandise sites like Amazon.com. Using this measured influence, we have proposed a method that evaluates loyalty of customers of a brand based on their ratings and sentiments of their reviews collected from online merchandise sites. Review network of a brand is built from all the reviews of all the products from that brand where nodes are customers and an edge is created if a customer becomes a potential reader of a review written by another customer. The centrality of a customer in that review network represents her influence. Our proposed method named as Social Promoter Score combines loyalty and centrality of all customers of a brand. We have compared our method with a baseline approach based on the concept of Net Promoter Score . We have applied Social Promoter Score on Amazon.com review data set of some well-known brands. Results show that Social Promoter Score predicts financial health of a brand in terms of future sales much better than baseline method. We have noticed that in general effects of Social Promoter Score reflect on the product sales in one to five months.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
