用高斯过程动态预测用户评分，比平均值更准10.2%。
Beyond Means: A Dynamic Framework for Predicting Customer Satisfaction
- 引入高斯过程建模评分随时间变化趋势
- 在Yelp数据上降低10.2%的预测误差
- 适合做在线评价系统优化的从业者
在线评分影响用户决策，但传统均值聚合方法无法适应质量随时间的变化，且忽略评论异质性（如情感倾向、有用性）。为此，我们展示高斯过程（GP）框架在评分聚合中的价值。提出一种定制化GP模型，既能捕捉评分随时间的动态变化，又能考虑评论异质性。基于来自Yelp的121,123条评分数据，对比不同聚合方法的预测能力，发现该GP模型显著更准确，相比样本均值将平均绝对误差降低10.2%。研究结果对营销实践者和消费者具有重要意义：通过超越均值，在线声誉系统设计者可呈现更具信息量与自适应性的聚合评分，成为客户满意度的精准信号。
原文摘要 · Abstract (English)
Online ratings influence customer decision-making, yet standard aggregation methods, such as the sample mean, fail to adapt to quality changes over time and ignore review heterogeneity (e.g., review sentiment, a review's helpfulness). To address these challenges, we demonstrate the value of using the Gaussian process (GP) framework for rating aggregation. Specifically, we present a tailored GP model that captures the dynamics of ratings over time while additionally accounting for review heterogeneity. Based on 121,123 ratings from Yelp, we compare the predictive power of different rating aggregation methods in predicting future ratings, thereby finding that the GP model is considerably more accurate and reduces the mean absolute error by 10.2% compared to the sample mean. Our findings have important implications for marketing practitioners and customers. By moving beyond means, designers of online reputation systems can display more informative and adaptive aggregated rating scores that are accurate signals of expected customer satisfaction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。