A Simple Deep Personalized Recommendation System
This work addresses the need for scalable and efficient personalized recommendations in two-sided vacation rental marketplaces, though it appears incremental as it builds on existing embedding and deep learning techniques.
The paper tackled the problem of learning user preferences from implicit signals in vacation rental marketplaces by proposing a Simple Deep Personalized Recommendation System that computes travelers' conditional embeddings. The result showed that traveler embeddings created using a Deep Average Network improved the precision of a downstream conversion prediction model by seven percent in offline evaluation.
Recommender systems are critical tools to match listings and travelers in two-sided vacation rental marketplaces. Such systems require high capacity to extract user preferences for items from implicit signals at scale. To learn those preferences, we propose a Simple Deep Personalized Recommendation System to compute travelers' conditional embeddings. Our method combines listing embeddings in a supervised structure to build short-term historical context to personalize recommendations for travelers. Deployed in the production environment, this approach is computationally efficient and scalable, and allows us to capture non-linear dependencies. Our offline evaluation indicates that traveler embeddings created using a Deep Average Network can improve the precision of a downstream conversion prediction model by seven percent, outperforming more complex benchmark methods for online shopping experience personalization.