Enhancing Sequential Music Recommendation with Personalized Popularity Awareness

TL;DR

The paper addresses the challenge that sequential music recommender systems, including Transformer-based models, often fail to capture the strong repeated listening signals inherent in music consumption. It introduces Personalized Popularity Awareness (PPS), which computes user-specific popularity signals from interaction counts $C$, with smoothed probabilities $\hat{p}_P(j)$, and conveys these signals to the model either by transforming them into logits $y_j$ for softmax-based methods or deriving compatible scores for sigmoid-based methods, thereby learning the delta from popularity. The key contributions include formalizing PPS, showing Personalized Most Popular as a strong baseline, and demonstrating that PPS-enhanced Transformer models achieve substantial gains (relative improvements from $25.2\%$ to $69.8\%$) on two large music datasets, bringing performance in line with or above the baseline. The work underscores the importance of modeling repetitive listening in music and provides a practical, available implementation to enable replication and further study in popularity-aware recommendation.

Abstract

In the realm of music recommendation, sequential recommender systems have shown promise in capturing the dynamic nature of music consumption. Nevertheless, traditional Transformer-based models, such as SASRec and BERT4Rec, while effective, encounter challenges due to the unique characteristics of music listening habits. In fact, existing models struggle to create a coherent listening experience due to rapidly evolving preferences. Moreover, music consumption is characterized by a prevalence of repeated listening, i.e., users frequently return to their favourite tracks, an important signal that could be framed as individual or personalized popularity. This paper addresses these challenges by introducing a novel approach that incorporates personalized popularity information into sequential recommendation. By combining user-item popularity scores with model-generated scores, our method effectively balances the exploration of new music with the satisfaction of user preferences. Experimental results demonstrate that a Personalized Most Popular recommender, a method solely based on user-specific popularity, outperforms existing state-of-the-art models. Furthermore, augmenting Transformer-based models with personalized popularity awareness yields superior performance, showing improvements ranging from 25.2% to 69.8%. The code for this paper is available at https://github.com/sisinflab/personalized-popularity-awareness.