Demystifying Core Ranking in Pinterest Image Search
This work addresses image search quality for hundreds of millions of Pinterest users, but it is incremental as it evolves existing ranking techniques.
The paper tackled improving Pinterest's image search quality by designing and deploying ranking pipelines, focusing on training data, featurization, and models, with extensive studies showing efficiency and effectiveness in the final launched models.
Pinterest Image Search Engine helps hundreds of millions of users discover interesting content everyday. This motivates us to improve the image search quality by evolving our ranking techniques. In this work, we share how we practically design and deploy various ranking pipelines into Pinterest image search ecosystem. Specifically, we focus on introducing our novel research and study on three aspects: training data, user/image featurization and ranking models. Extensive offline and online studies compared the performance of different models and demonstrated the efficiency and effectiveness of our final launched ranking models.