Deep Session Interest Network for Click-Through Rate Prediction

Abstract

Click-Through Rate (CTR) prediction plays an important role in many industrial applications, such as online advertising and recommender systems. How to capture users' dynamic and evolving interests from their behavior sequences remains a continuous research topic in the CTR prediction. However, most existing studies overlook the intrinsic structure of the sequences: the sequences are composed of sessions, where sessions are user behaviors separated by their occurring time. We observe that user behaviors are highly homogeneous in each session, and heterogeneous cross sessions. Based on this observation, we propose a novel CTR model named Deep Session Interest Network (DSIN) that leverages users' multiple historical sessions in their behavior sequences. We first use self-attention mechanism with bias encoding to extract users' interests in each session. Then we apply Bi-LSTM to model how users' interests evolve and interact among sessions. Finally, we employ the local activation unit to adaptively learn the influences of various session interests on the target item. Experiments are conducted on both advertising and production recommender datasets and DSIN outperforms other state-of-the-art models on both datasets.

Has been accepted by IJCAI-2019, 7 pages, 3 pages, 3 pics

References in corpus (6)

• Session-based Recommendations with Recurrent Neural Networks
• DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
• Wide & Deep Learning for Recommender Systems
• Learning from History and Present: Next-item Recommendation via Discriminatively Exploiting User Behaviors
• Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks
• Self-Attentive Sequential Recommendation

Cited by in corpus (21)

• SAR-Net: A Scenario-Aware Ranking Network for Personalized Fair Recommendation in Hundreds of Travel Scenarios
• Deep Position-wise Interaction Network for CTR Prediction
• End-to-End User Behavior Retrieval in Click-Through RatePrediction Model
• SDM: Sequential Deep Matching Model for Online Large-scale Recommender System
• Entire Space Multi-Task Modeling via Post-Click Behavior Decomposition for Conversion Rate Prediction
• Kalman Filtering Attention for User Behavior Modeling in CTR Prediction
• Pairwise Interactive Graph Attention Network for Context-Aware Recommendation
• FedCTR: Federated Native Ad CTR Prediction with Multi-Platform User Behavior Data
• Search-based User Interest Modeling with Lifelong Sequential Behavior Data for Click-Through Rate Prediction
• DAGNN: Demand-aware Graph Neural Networks for Session-based Recommendation
• Denoising User-aware Memory Network for Recommendation
• DemiNet: Dependency-Aware Multi-Interest Network with Self-Supervised Graph Learning for Click-Through Rate Prediction
• Dynamic Sequential Graph Learning for Click-Through Rate Prediction
• Hierarchically Modeling Micro and Macro Behaviors via Multi-Task Learning for Conversion Rate Prediction
• Scenario-aware and Mutual-based approach for Multi-scenario Recommendation in E-Commerce
• Conceptualize and Infer User Needs in E-commerce
• PURS: Personalized Unexpected Recommender System for Improving User Satisfaction
• Embedding-based Product Retrieval in Taobao Search
• AdnFM: An Attentive DenseNet based Factorization Machine for CTR Prediction
• Dual Graph enhanced Embedding Neural Network for CTR Prediction
• Dual Attentive Sequential Learning for Cross-Domain Click-Through Rate Prediction