1. NCCU Academic Hub
  2. 學術產出
  3. 資訊學院
  4. 會議論文
  5. Item

學術產出-會議論文

文章檢視/開啟

書目匯出

Google ScholarTM

政大圖書館

引文資訊

TAIR相關學術產出

題名 Leverage Item Popularity and Recommendation Quality via Cost-sensitive Factorization Machines
作者 Chen, Chih-Ming;Chen, Hsin-Ping;Tsai, Ming-Feng;Yang, Yi-Hsuan
陳志明;陳心蘋;蔡銘峰
貢獻者 經濟系;資科系
日期 2014-12
上傳時間 22-六月-2016 17:19:59 (UTC+8)
摘要 The accuracy of recommendation trends to be worse towards the long tail of the popularity distribution of items, but items in the long tail are generally considered to be valuable as they occupy a majority part of entire data. In this paper, we develop an instance-level cost-sensitive Factorization Machine (FM) to tackle the problem. The new algorithm allows the FM model to automatically leverage the trade-off between item popularity and recommendation quality. Specifically, by adding a cost criterion to the loss function, the FM model is now able to discriminate the relative importance of popularity from massive data. In addition, we convert several well-known functions into the popularity weighting functions, thereby demonstrating that the proposed method can fit the model parameters to various kinds of measurements. In the experiments, we assess the performance on a real-world music dataset which is collected from an online music streaming service, KKBOX. The dataset contains 1,800,000 listening records that cover 5,000 users and 30,000 songs. The results show that, the proposed method not only keeps the performance as primitive model but also avoids retrieving too much popular music in the top recommendations.
關聯 Proceedings of the 2014 IEEE International Conference on Data Mining (ICDM `14), 1158-1162, 2014
國際標準書號9781479942756
資料類型 conference
dc.contributor 經濟系;資科系-
dc.creator (作者) Chen, Chih-Ming;Chen, Hsin-Ping;Tsai, Ming-Feng;Yang, Yi-Hsuan-
dc.creator (作者) 陳志明;陳心蘋;蔡銘峰zh_TW
dc.date (日期) 2014-12-
dc.date.accessioned 22-六月-2016 17:19:59 (UTC+8)-
dc.date.available 22-六月-2016 17:19:59 (UTC+8)-
dc.date.issued (上傳時間) 22-六月-2016 17:19:59 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/98247-
dc.description.abstract (摘要) The accuracy of recommendation trends to be worse towards the long tail of the popularity distribution of items, but items in the long tail are generally considered to be valuable as they occupy a majority part of entire data. In this paper, we develop an instance-level cost-sensitive Factorization Machine (FM) to tackle the problem. The new algorithm allows the FM model to automatically leverage the trade-off between item popularity and recommendation quality. Specifically, by adding a cost criterion to the loss function, the FM model is now able to discriminate the relative importance of popularity from massive data. In addition, we convert several well-known functions into the popularity weighting functions, thereby demonstrating that the proposed method can fit the model parameters to various kinds of measurements. In the experiments, we assess the performance on a real-world music dataset which is collected from an online music streaming service, KKBOX. The dataset contains 1,800,000 listening records that cover 5,000 users and 30,000 songs. The results show that, the proposed method not only keeps the performance as primitive model but also avoids retrieving too much popular music in the top recommendations.-
dc.format.extent 126 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) Proceedings of the 2014 IEEE International Conference on Data Mining (ICDM `14), 1158-1162, 2014-
dc.relation (關聯) 國際標準書號9781479942756-
dc.title (題名) Leverage Item Popularity and Recommendation Quality via Cost-sensitive Factorization Machines-
dc.type (資料類型) conference-