Multimodal Enhancement of Sequential Recommendation

Researchers developed MuSTRec, a new AI framework that combines text and visual features to improve product recommendations. It builds item graphs and uses a frequency-based attention module to understand both short- and long-term user preferences. Tests on Amazon datasets show it outperforms current state-of-the-art models by up to 33.5%. The method also showed that integrating user data can improve short-term metrics by up to 200% on smaller datasets.