This paper describes the approach of the D2KLab team to the RecSys Challenge 2018 that focuses on the task of playlist completion. We propose an ensemble strategy of different recurrent neural networks leveraging pre-trained embeddings representing tracks, artists, albums, and titles as inputs. We also use lyrics from which we extract semantic and stylistic features that we fed into the network for the creative track. The RNN learns a probabilistic model from the sequences of items in the playlist, which is then used to predict the most likely tracks to be added to the playlist. Concerning the playlists without tracks, we implemented a fall-back strategy called Title2Rec that generates recommendations using only the playlist title. We optimized the RNN, Title2Rec, and the ensemble approach on a validation set, tuning hyper-parameters such as the optimizer algorithm, the learning rate, and the generation strategy. This approach is effective in predicting tracks for a playlist and flexible to include diverse types of inputs, but it is also computationally demanding in the training phase.

An ensemble approach of recurrent neural networks using pre-trained embeddings for playlist completion / Monti, DIEGO MICHELE; Palumbo, Enrico; Rizzo, Giuseppe; Lisena, Pasquale; Troncy, Raphaël; Fell, Michael; Cabrio, Elena; Morisio, Maurizio. - ELETTRONICO. - (2018), pp. 131-136. ((Intervento presentato al convegno 12th ACM Conference on Recommender Systems tenutosi a Vancouver (CA) nel October 2, 2018 [10.1145/3267471.3267484].

An ensemble approach of recurrent neural networks using pre-trained embeddings for playlist completion

Diego Monti;Enrico Palumbo;Giuseppe Rizzo;LISENA, PASQUALE;Maurizio Morisio
2018

Abstract

This paper describes the approach of the D2KLab team to the RecSys Challenge 2018 that focuses on the task of playlist completion. We propose an ensemble strategy of different recurrent neural networks leveraging pre-trained embeddings representing tracks, artists, albums, and titles as inputs. We also use lyrics from which we extract semantic and stylistic features that we fed into the network for the creative track. The RNN learns a probabilistic model from the sequences of items in the playlist, which is then used to predict the most likely tracks to be added to the playlist. Concerning the playlists without tracks, we implemented a fall-back strategy called Title2Rec that generates recommendations using only the playlist title. We optimized the RNN, Title2Rec, and the ensemble approach on a validation set, tuning hyper-parameters such as the optimizer algorithm, the learning rate, and the generation strategy. This approach is effective in predicting tracks for a playlist and flexible to include diverse types of inputs, but it is also computationally demanding in the training phase.
978-1-4503-6586-4
File in questo prodotto:
File Dimensione Formato  
RecSys_Challenge_2018.pdf

accesso aperto

Descrizione: Articolo principale
Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 676.47 kB
Formato Adobe PDF
676.47 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2712365
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo