Improving Cold-start Item Advertisement for Small Businesses
In this paper, we study cold-start item advertisements for small businesses on a real-world E-commerce website. From analysis, we found that the existing cold-start Recommender Systems (RSs) are not helpful for small businesses with a few sales history. Training samples in RS models can be extremely biased towards popular items or shops with sufficient sales history, and can decrease advertising performance for small shops with few or zero sales history. We propose two solutions to improve advertising performance for small shops: negative sampling and Meta-shop. Negative sampling focuses on changing the data distribution and Meta-shop focuses on building novel meta-learning models. By including sales information in the training of both methods, we are able to learn better cold-start item representations from small shops while keeping the same or better overall recommendation performance.We conducted experiments on a real-world E-commerce dataset and demonstrated that the proposed methods outperformed a production baseline. Specifically, we achieved up to 19.6% relative improvement of Recall@10k using Meta-shop compared to the traditional cold-start RS model.