Search & Recommendations

Search & recommendations research at Spotify focuses on identifying ways to provide users with seamless access to their favorite audio content, from music to podcasts, and to help them explore their taste. A big focus is on building highly effective, personalized and interactive models that exploit contextual information and historical user interactions. Our research interests include large-scale recommendation algorithms for music and more generally audio discovery, algorithms for audio search through voice and text, query analysis for effective query suggestion, query completion and search assistance, multilingual information retrieval for voice search, and ranking algorithm for revenue and music ads.

Latest Search & Recommendations Publications

November 2021 | ISMIR - International Society for Music Information Retrieval Conference

Multi-Task Learning of Graph-based Inductive Representations of Music Content

Antonia Saravanou, Federico Tomasi, Rishabh Mehrotra and Mounia Lalmas

November 2021 | CIKM

Leveraging Semantic Information to Facilitate the Discovery of Underserved Podcasts

Maryam Aziz, Alice Wang, Aasish Pappu, Hugues Bouchard,Yu Zhao, Benjamin Carterette and Mounia Lalmas

Other Research Areas