Abstract:Click-through Rate (CTR) prediction is crucial for online personalization platforms. Recent advancements have shown that modeling rich user behaviors can significantly improve the performance of CTR prediction. Current long-term user behavior modeling algorithms predominantly follow two cascading stages. The first stage retrieves subsequence related to the target item from the long-term behavior sequence, while the second stage models the relationship between the subsequence and the target item. Despite significant progress, these methods have two critical flaws. First, the retrieval query typically includes only target item information, limiting the ability to capture the user's diverse interests. Second, relational information, such as sequential and interactive information within the subsequence, is frequently overlooked. Therefore, it requires to be further mined to more accurately model user interests. To this end, we propose Multi-granularity Interest Retrieval and Refinement Network (MIRRN). Specifically, we first construct queries based on behaviors observed at different time scales to obtain subsequences, each capturing users' interest at various granularities. We then introduce an noval multi-head Fourier transformer to efficiently learn sequential and interactive information within the subsequences, leading to more accurate modeling of user interests. Finally, we employ multi-head target attention to adaptively assess the impact of these multi-granularity interests on the target item. Extensive experiments have demonstrated that MIRRN significantly outperforms state-of-the-art baselines. Furthermore, an A/B test shows that MIRRN increases the average number of listening songs by 1.32% and the average time of listening songs by 0.55% on a popular music streaming app. The implementation code is publicly available at <a class="link-external link-https" href="https://github.com/psycho-demon/MIRRN" rel="external noopener nofollow">this https URL</a>.

Modeling the Interest-Forgetting Curve for Music Recommendation

Recommending More Suitable Music Based on Users' Real Context.

A Personalized Interest-Forgetting Markov Model For Recommendations

Sequential Modeling of Hierarchical User Intention and Preference for Next-item Recommendation

Learning to Embed Music and Metadata for Context-Aware Music Recommendation

Sequence-based Context-Aware Music Recommendation

Music Recommendation Via Heterogeneous Information Graph Embedding.

Learning Music Embedding with Metadata for Context Aware Recommendation

Modeling Sequential Listening Behaviors with Attentive Temporal Point Process for Next and Next New Music Recommendation

Local Implicit Feedback Mining for Music Recommendation

Time-Aware Music Recommender Systems: Modeling the Evolution of Implicit User Preferences and User Listening Habits in A Collaborative Filtering Approach

Music Recommendation Using Graph Based Quality Model

Design of Deep Learning Network Model for Personalized Music Emotional Recommendation

The Potential of the Confluence of Theoretical and Algorithmic Modeling in Music Recommendation

Smart Recommendation of Cloud Music Services Based on Contextualized Mentality Modeling and Beyond

A Deep Temporal Neural Music Recommendation Model Utilizing Music and User Metadata

STAMP: Short-Term Attention/Memory Priority Model for Session-based Recommendation.

Interest Clock: Time Perception in Real-Time Streaming Recommendation System

Smart Recommendation of Personalized Cloud Music Service Based on Mental Model.

Leveraging the structure of musical preference in content-aware music recommendation

Multi-granularity Interest Retrieval and Refinement Network for Long-Term User Behavior Modeling in CTR Prediction