Towards Seed-Free Music Playlist Generation: Enhancing collaborative Filtering with Playlist Title Information

Jaehun Kim; Minz Won; Cynthia C.S. Liem; Alan Hanjalic

doi:10.1145/3267471.3267485

Towards Seed-Free Music Playlist Generation: Enhancing collaborative Filtering with Playlist Title Information

Jaehun Kim, Minz Won, Cynthia C.S. Liem, Alan Hanjalic

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

7 Citations (Scopus)

70 Downloads (Pure)

Abstract

In this paper, we propose a hybrid Neural Collaborative Filtering (NCF) model trained with a multi-objective function to achieve a music playlist generation system. The proposed approach focuses particularly on the cold-start problem (playlists with no seed tracks) and uses a text encoder employing a Recurrent Neural Network (RNN) to exploit textual information given by the playlist title. To accelerate the training, we first apply Weighted Regularized Matrix Factorization (WRMF) as the basic recommendation model to prelearn latent factors of playlists and tracks. These factors then feed into the proposed multi-objective optimization that also involves embeddings of playlist titles. The experimental study indicates that the proposed approach can effectively suggest suitable music tracks for a given playlist title, compensating poor original recommendation results made on empty playlists by the WRMF model.

Original language	English
Title of host publication	RecSys Challenge '18
Subtitle of host publication	Proceedings of the ACM Recommender Systems Challenge 2018
Place of Publication	New York
Publisher	Association for Computing Machinery (ACM)
Pages	1-6
Number of pages	6
ISBN (Electronic)	978-1-4503-6586-4
DOIs	https://doi.org/10.1145/3267471.3267485
Publication status	Published - 2018
Event	Workshop on Recommenders in Tourism co-located with the 12th ACM Conference on Recommender Systems (RecSys 2018) - Vancouver, Canada Duration: 7 Oct 2018 → 7 Oct 2018

Workshop

Workshop	Workshop on Recommenders in Tourism co-located with the 12th ACM Conference on Recommender Systems (RecSys 2018)
Country/Territory	Canada
City	Vancouver
Period	7/10/18 → 7/10/18

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Collaborative filtering
Hybrid recommender system
LSTM
Multi-objective function
Music playlist generation
WRMF

Access to Document

10.1145/3267471.3267485

a14-KimFinal published version, 848 KB

Cite this

@inproceedings{ec345852ae51436cb87a63b468bbf15a,

title = "Towards Seed-Free Music Playlist Generation: Enhancing collaborative Filtering with Playlist Title Information",

abstract = "In this paper, we propose a hybrid Neural Collaborative Filtering (NCF) model trained with a multi-objective function to achieve a music playlist generation system. The proposed approach focuses particularly on the cold-start problem (playlists with no seed tracks) and uses a text encoder employing a Recurrent Neural Network (RNN) to exploit textual information given by the playlist title. To accelerate the training, we first apply Weighted Regularized Matrix Factorization (WRMF) as the basic recommendation model to prelearn latent factors of playlists and tracks. These factors then feed into the proposed multi-objective optimization that also involves embeddings of playlist titles. The experimental study indicates that the proposed approach can effectively suggest suitable music tracks for a given playlist title, compensating poor original recommendation results made on empty playlists by the WRMF model.",

keywords = "Collaborative filtering, Hybrid recommender system, LSTM, Multi-objective function, Music playlist generation, WRMF",

author = "Jaehun Kim and Minz Won and Liem, {Cynthia C.S.} and Alan Hanjalic",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; Workshop on Recommenders in Tourism<br/>co-located with the 12th ACM Conference on Recommender Systems (RecSys 2018) ; Conference date: 07-10-2018 Through 07-10-2018",

year = "2018",

doi = "10.1145/3267471.3267485",

language = "English",

pages = "1--6",

booktitle = "RecSys Challenge '18",

publisher = "Association for Computing Machinery (ACM)",

address = "United States",

}

Kim, J, Won, M, Liem, CCS & Hanjalic, A 2018, Towards Seed-Free Music Playlist Generation: Enhancing collaborative Filtering with Playlist Title Information. in RecSys Challenge '18 : Proceedings of the ACM Recommender Systems Challenge 2018 ., a14, Association for Computing Machinery (ACM), New York, pp. 1-6, Workshop on Recommenders in Tourism
co-located with the 12th ACM Conference on Recommender Systems (RecSys 2018), Vancouver, Canada, 7/10/18. https://doi.org/10.1145/3267471.3267485

Towards Seed-Free Music Playlist Generation: Enhancing collaborative Filtering with Playlist Title Information. / Kim, Jaehun; Won, Minz; Liem, Cynthia C.S. et al.
RecSys Challenge '18 : Proceedings of the ACM Recommender Systems Challenge 2018 . New York: Association for Computing Machinery (ACM), 2018. p. 1-6 a14.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Towards Seed-Free Music Playlist Generation

T2 - Workshop on Recommenders in Tourism<br/>co-located with the 12th ACM Conference on Recommender Systems (RecSys 2018)

AU - Kim, Jaehun

AU - Won, Minz

AU - Liem, Cynthia C.S.

AU - Hanjalic, Alan

N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2018

Y1 - 2018

N2 - In this paper, we propose a hybrid Neural Collaborative Filtering (NCF) model trained with a multi-objective function to achieve a music playlist generation system. The proposed approach focuses particularly on the cold-start problem (playlists with no seed tracks) and uses a text encoder employing a Recurrent Neural Network (RNN) to exploit textual information given by the playlist title. To accelerate the training, we first apply Weighted Regularized Matrix Factorization (WRMF) as the basic recommendation model to prelearn latent factors of playlists and tracks. These factors then feed into the proposed multi-objective optimization that also involves embeddings of playlist titles. The experimental study indicates that the proposed approach can effectively suggest suitable music tracks for a given playlist title, compensating poor original recommendation results made on empty playlists by the WRMF model.

AB - In this paper, we propose a hybrid Neural Collaborative Filtering (NCF) model trained with a multi-objective function to achieve a music playlist generation system. The proposed approach focuses particularly on the cold-start problem (playlists with no seed tracks) and uses a text encoder employing a Recurrent Neural Network (RNN) to exploit textual information given by the playlist title. To accelerate the training, we first apply Weighted Regularized Matrix Factorization (WRMF) as the basic recommendation model to prelearn latent factors of playlists and tracks. These factors then feed into the proposed multi-objective optimization that also involves embeddings of playlist titles. The experimental study indicates that the proposed approach can effectively suggest suitable music tracks for a given playlist title, compensating poor original recommendation results made on empty playlists by the WRMF model.

KW - Collaborative filtering

KW - Hybrid recommender system

KW - LSTM

KW - Multi-objective function

KW - Music playlist generation

KW - WRMF

UR - http://www.scopus.com/inward/record.url?scp=85056704269&partnerID=8YFLogxK

U2 - 10.1145/3267471.3267485

DO - 10.1145/3267471.3267485

M3 - Conference contribution

SP - 1

EP - 6

BT - RecSys Challenge '18

PB - Association for Computing Machinery (ACM)

CY - New York

Y2 - 7 October 2018 through 7 October 2018

ER -

Towards Seed-Free Music Playlist Generation: Enhancing collaborative Filtering with Playlist Title Information

Abstract

Workshop

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this