Classical music for rock fans? Novel recommendations for expanding user interests

Makoto Nakatsuji, Yasuhiro Fujiwara, Akimichi Tanaka, Toshio Uchiyama, Ko Fujimura, Toru Ishida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

33 Citations (Scopus)

Abstract

Most recommender algorithms produce types similar to those the active user has accessed before. This is because they measure user similarity only from the co-rating behaviors against items and compute recommendations by analyzing the items possessed by the users most similar to the active user. In this paper, we define item novelty as the smallest distance from the class the user accessed before to the class that includes target items over the taxonomy. Then, we try to accurately recommend highly novel items to the user. First, our method measures user similarity by employing items rated by users and a taxonomy of items. It can accurately identify many items that may suit the user. Second, it creates a graph whose nodes are users; weighted edges are set between users according to their similarity. It analyzes the user graph and extracts users that are related on the graph though the similarity between the active user and each of those users is not high. The users so extracted are likely to have highly novel items for the active user. An evaluation conducted on several datasets finds that our method accurately identifies items with higher novelty than previous methods.

Original languageEnglish
Title of host publicationCIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops
Pages949-958
Number of pages10
DOIs
Publication statusPublished - 2010 Dec 1
Externally publishedYes
Event19th International Conference on Information and Knowledge Management and Co-located Workshops, CIKM'10 - Toronto, ON, Canada
Duration: 2010 Oct 262010 Oct 30

Publication series

NameInternational Conference on Information and Knowledge Management, Proceedings

Conference

Conference19th International Conference on Information and Knowledge Management and Co-located Workshops, CIKM'10
CountryCanada
CityToronto, ON
Period10/10/2610/10/30

Fingerprint

Music
Graph
Taxonomy
Novelty
Evaluation
Node
Rating

Keywords

  • Collaborative filtering
  • Novelty detection

ASJC Scopus subject areas

  • Decision Sciences(all)
  • Business, Management and Accounting(all)

Cite this

Nakatsuji, M., Fujiwara, Y., Tanaka, A., Uchiyama, T., Fujimura, K., & Ishida, T. (2010). Classical music for rock fans? Novel recommendations for expanding user interests. In CIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops (pp. 949-958). (International Conference on Information and Knowledge Management, Proceedings). https://doi.org/10.1145/1871437.1871558

Classical music for rock fans? Novel recommendations for expanding user interests. / Nakatsuji, Makoto; Fujiwara, Yasuhiro; Tanaka, Akimichi; Uchiyama, Toshio; Fujimura, Ko; Ishida, Toru.

CIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops. 2010. p. 949-958 (International Conference on Information and Knowledge Management, Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nakatsuji, M, Fujiwara, Y, Tanaka, A, Uchiyama, T, Fujimura, K & Ishida, T 2010, Classical music for rock fans? Novel recommendations for expanding user interests. in CIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops. International Conference on Information and Knowledge Management, Proceedings, pp. 949-958, 19th International Conference on Information and Knowledge Management and Co-located Workshops, CIKM'10, Toronto, ON, Canada, 10/10/26. https://doi.org/10.1145/1871437.1871558
Nakatsuji M, Fujiwara Y, Tanaka A, Uchiyama T, Fujimura K, Ishida T. Classical music for rock fans? Novel recommendations for expanding user interests. In CIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops. 2010. p. 949-958. (International Conference on Information and Knowledge Management, Proceedings). https://doi.org/10.1145/1871437.1871558
Nakatsuji, Makoto ; Fujiwara, Yasuhiro ; Tanaka, Akimichi ; Uchiyama, Toshio ; Fujimura, Ko ; Ishida, Toru. / Classical music for rock fans? Novel recommendations for expanding user interests. CIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops. 2010. pp. 949-958 (International Conference on Information and Knowledge Management, Proceedings).
@inproceedings{440c253c0c574e629a392df4985d31c5,
title = "Classical music for rock fans?: Novel recommendations for expanding user interests",
abstract = "Most recommender algorithms produce types similar to those the active user has accessed before. This is because they measure user similarity only from the co-rating behaviors against items and compute recommendations by analyzing the items possessed by the users most similar to the active user. In this paper, we define item novelty as the smallest distance from the class the user accessed before to the class that includes target items over the taxonomy. Then, we try to accurately recommend highly novel items to the user. First, our method measures user similarity by employing items rated by users and a taxonomy of items. It can accurately identify many items that may suit the user. Second, it creates a graph whose nodes are users; weighted edges are set between users according to their similarity. It analyzes the user graph and extracts users that are related on the graph though the similarity between the active user and each of those users is not high. The users so extracted are likely to have highly novel items for the active user. An evaluation conducted on several datasets finds that our method accurately identifies items with higher novelty than previous methods.",
keywords = "Collaborative filtering, Novelty detection",
author = "Makoto Nakatsuji and Yasuhiro Fujiwara and Akimichi Tanaka and Toshio Uchiyama and Ko Fujimura and Toru Ishida",
year = "2010",
month = "12",
day = "1",
doi = "10.1145/1871437.1871558",
language = "English",
isbn = "9781450300995",
series = "International Conference on Information and Knowledge Management, Proceedings",
pages = "949--958",
booktitle = "CIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops",

}

TY - GEN

T1 - Classical music for rock fans?

T2 - Novel recommendations for expanding user interests

AU - Nakatsuji, Makoto

AU - Fujiwara, Yasuhiro

AU - Tanaka, Akimichi

AU - Uchiyama, Toshio

AU - Fujimura, Ko

AU - Ishida, Toru

PY - 2010/12/1

Y1 - 2010/12/1

N2 - Most recommender algorithms produce types similar to those the active user has accessed before. This is because they measure user similarity only from the co-rating behaviors against items and compute recommendations by analyzing the items possessed by the users most similar to the active user. In this paper, we define item novelty as the smallest distance from the class the user accessed before to the class that includes target items over the taxonomy. Then, we try to accurately recommend highly novel items to the user. First, our method measures user similarity by employing items rated by users and a taxonomy of items. It can accurately identify many items that may suit the user. Second, it creates a graph whose nodes are users; weighted edges are set between users according to their similarity. It analyzes the user graph and extracts users that are related on the graph though the similarity between the active user and each of those users is not high. The users so extracted are likely to have highly novel items for the active user. An evaluation conducted on several datasets finds that our method accurately identifies items with higher novelty than previous methods.

AB - Most recommender algorithms produce types similar to those the active user has accessed before. This is because they measure user similarity only from the co-rating behaviors against items and compute recommendations by analyzing the items possessed by the users most similar to the active user. In this paper, we define item novelty as the smallest distance from the class the user accessed before to the class that includes target items over the taxonomy. Then, we try to accurately recommend highly novel items to the user. First, our method measures user similarity by employing items rated by users and a taxonomy of items. It can accurately identify many items that may suit the user. Second, it creates a graph whose nodes are users; weighted edges are set between users according to their similarity. It analyzes the user graph and extracts users that are related on the graph though the similarity between the active user and each of those users is not high. The users so extracted are likely to have highly novel items for the active user. An evaluation conducted on several datasets finds that our method accurately identifies items with higher novelty than previous methods.

KW - Collaborative filtering

KW - Novelty detection

UR - http://www.scopus.com/inward/record.url?scp=78651316651&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651316651&partnerID=8YFLogxK

U2 - 10.1145/1871437.1871558

DO - 10.1145/1871437.1871558

M3 - Conference contribution

AN - SCOPUS:78651316651

SN - 9781450300995

T3 - International Conference on Information and Knowledge Management, Proceedings

SP - 949

EP - 958

BT - CIKM'10 - Proceedings of the 19th International Conference on Information and Knowledge Management and Co-located Workshops

ER -