Reading the Mood: Emotion-Guided Book-to-Music Recommendation via CGANs and LLMs

2026-10-05

AI Summary

This study addresses the challenge of cross-domain sentiment matching and personalized recommendation between books and background music to enhance reading immersion. We propose SAGA-CDR, a novel two-stage cross-domain recommendation framework. The first stage leverages large language models to classify books into sentiment quadrants for candidate track filtering. The second stage employs a masked conditional generative adversarial network (CGAN) to map Transformer-based sentiment embeddings derived from user reviews into the music domain, effectively handling missing data and predicting ratings. Experimental results demonstrate that SAGA-CDR achieves state-of-the-art accuracy on both Amazon and Douban datasets, yielding RMSE scores of 0.98 and 0.91, respectively. Furthermore, its ranking performance is comparable to the strongest baselines while exhibiting significant cross-lingual generalization capabilities.

Abstract

Background music that matches the mood of a text has been shown to make readers feel more immersed and improve their reading experience, motivating recommender systems that pair books with mood-matched music. In this direction, we present Sentiment Aware Generative Adversarial Network for Cross Domain Recommendation (SAGA-CDR), a two-phase cross-domain recommendation framework that personalizes music suggestions and emotionally aligns them with the book being read. In the first phase, transformer-based sentiment embeddings are constructed from user reviews and mapped across domains via a Conditional Generative Adversarial Network, whose mask-conditioned generator handles missing sentiment components and injects stochasticity for richer preference transfer. A compact rating neural network then fuses sentiment-specific interaction scores with a collaborative filtering prior to predict music ratings. In the second phase, large language models classify each book into a valence-arousal emotional quadrant, and candidate tracks are filtered to match that quadrant. Experiments on both the English Amazon and Chinese Douban datasets show that SAGA-CDR achieves the best rating prediction accuracy on Amazon (RMSE 0.98) and the lowest RMSE on Douban (0.91), with ranking performance competitive with the strongest sentiment-aware baseline, even in cross-lingual settings.

Problem

• cross-domain recommendation
• book-to-music recommendation
• emotion alignment
• sentiment-aware recommendation

Innovation

• Cross-Domain Recommendation
• Conditional GAN
• Large Language Models
• Sentiment Embeddings
• Emotion-Guided Filtering