Matching novelty while training: Novel recommendation based on personalized pairwise loss weighting

Kachun Lo, Tsukasa Ishigaki

研究成果: Conference contribution

抄録

Most works of recommender system seek to provide highly accurate item prediction while having potentially great bias to popular items. Both users and items' providers will suffer if their system has strong preference for monotonous popular items. A better system should consider also item novelty. Previous works of novel recommendation focus mainly on re-ranking a top-N list generated by an accuracy-focused base model. As a result, these frameworks are 2-stage and essentially limited to the base model. In addition, when training the base model, the common BRP loss function treats all pairs in the same manner, consistently suppresses interesting negative items which should have been recommended. In this work, we propose a personalized pairwise novelty weighting for BPR loss function, which covers the limitations of BPR and effectively improves novelty with marginal loss in accuracy. Base model will be guided by the loss weights to learn user preference and to generate novel suggestion list in 1 stage. Comprehensive experiments on 3 public datasets show that our approach effectively promotes novelty with almost no decrease in accuracy.

本文言語English
ホスト出版物のタイトルProceedings - 19th IEEE International Conference on Data Mining, ICDM 2019
編集者Jianyong Wang, Kyuseok Shim, Xindong Wu
出版社Institute of Electrical and Electronics Engineers Inc.
ページ468-477
ページ数10
ISBN(電子版)9781728146034
DOI
出版ステータスPublished - 2019 11
イベント19th IEEE International Conference on Data Mining, ICDM 2019 - Beijing, China
継続期間: 2019 11 82019 11 11

出版物シリーズ

名前Proceedings - IEEE International Conference on Data Mining, ICDM
2019-November
ISSN(印刷版)1550-4786

Conference

Conference19th IEEE International Conference on Data Mining, ICDM 2019
CountryChina
CityBeijing
Period19/11/819/11/11

ASJC Scopus subject areas

  • Engineering(all)

フィンガープリント 「Matching novelty while training: Novel recommendation based on personalized pairwise loss weighting」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル