ColdGuess: A general and effective relational graph convolutional network to tackle cold start cases

Low-quality listings and bad actor behavior in online retail websites threatens e-commerce business as these result in sub-optimal buying experience and erode customer trust. When a new listing is created, how to tell it has good quality? Is the method effective, fast, and scalable? Previous approaches often face three limitations/challenges: (1) unable to handle cold start problems where new sellers/listings lack sufficient selling histories. (2) inability of scoring hundreds of millions of listings at scale, or compromise performance for scalability. (3) has space challenge from large-scale graph built on giant business size. To overcome these limitations, we proposed ColdGuess, an inductive graph-based risk detector built upon a heterogeneous seller-product graph, which effectively identifies risky seller/product /listings at scale. ColdGuess tackles the large-scale graph by consolidated nodes, and addresses the cold start problem using homogeneous influence1. The evaluation on real data demonstrates that ColdGuess has stable performance as the number of unknown features increases. It outperforms the lightgbm2, a commonly used risk detection model in production, by up to 34 pcp ROC AUC in the cold start case when a new seller sells a new product . The resulting system, ColdGuess, is effective, adaptable to changing bad actor behavior, and is already in production. This paper belongs to “Application and analysis – Large-scale graph and modeling”, and in the "Novel research paper" category.