Deeply Supervised Semantic Model for Click-Through Rate Prediction in Sponsored Search
This work addresses the challenge of accurately predicting ad clicks and relevance in commercial search engines, offering incremental improvements over existing deep learning methods.
The paper tackles the problem of improving click-through rate prediction and query-ad matching in sponsored search by proposing a deeply supervised architecture that jointly learns semantic embeddings and CTR, achieving a 2% AUC improvement for CTR prediction and a 0.5% NDCG gain for matching.
In sponsored search it is critical to match ads that are relevant to a query and to accurately predict their likelihood of being clicked. Commercial search engines typically use machine learning models for both query-ad relevance matching and click-through-rate (CTR) prediction. However, matching models are based on the similarity between a query and an ad, ignoring the fact that a retrieved ad may not attract clicks, while click models rely on click history, being of limited use for new queries and ads. We propose a deeply supervised architecture that jointly learns the semantic embeddings of a query and an ad as well as their corresponding CTR.We also propose a novel cohort negative sampling technique for learning implicit negative signals. We trained the proposed architecture using one billion query-ad pairs from a major commercial web search engine. This architecture improves the best-performing baseline deep neural architectures by 2\% of AUC for CTR prediction and by statistically significant 0.5\% of NDCG for query-ad matching.