Deep neural network marketplace recommenders in online experiments

TL;DR

This paper explores the use of various deep neural network recommenders in a real-world online marketplace, demonstrating their effectiveness in improving user engagement and personalization.

Contribution

It introduces hybrid, sequence-based, and multi-armed bandit neural recommenders and evaluates their performance through online experiments in a production environment.

Findings

Hybrid models improve relevance by combining user engagement and content features

Sequence-based models capture user behavior patterns effectively

Multi-armed bandit models optimize engagement through dynamic re-ranking

Abstract

Recommendations are broadly used in marketplaces to match users with items relevant to their interests and needs. To understand user intent and tailor recommendations to their needs, we use deep learning to explore various heterogeneous data available in marketplaces. This paper focuses on the challenge of measuring recommender performance and summarizes the online experiment results with several promising types of deep neural network recommenders - hybrid item representation models combining features from user engagement and content, sequence-based models, and multi-armed bandit models that optimize user engagement by re-ranking proposals from multiple submodels. The recommenders are currently running in production at the leading Norwegian marketplace FINN.no and serves over one million visitors everyday.