作者: Çağatay Demiralp , Wang-Chiew Tan , Sara Evensen , Yuliang Li , Jonathan Engel
DOI:
关键词:
摘要: Reviews are integral to e-commerce services and products. They contain a wealth of information about the opinions experiences users, which can help better understand consumer decisions improve user experience with products services. Today, data scientists analyze reviews by developing rules models extract, aggregate, embedded in review text. However, working thousands reviews, typically noisy incomplete text, be daunting without proper tools. Here we first contribute results from an interview study that conducted fifteen who work providing insights into their practices challenges. Results suggest need interactive systems for many analysis tasks. In response introduce Teddy, system enables quickly obtain extraction modeling pipelines.