Publications-Periodical Articles

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 Improving Ranking Performance with Cost-sensitive Ordinal Classification via Regression
作者 Ruan, Yu-Xun ; Lin, Hsuan-Tien ; Tsai, Ming-Feng
蔡銘峰
貢獻者 資科系
關鍵詞 List-wise ranking ; Cost-sensitive ; Regression ; Reduction
日期 2014.02
上傳時間 6-Mar-2014 16:29:52 (UTC+8)
摘要 This paper proposes a novel ranking approach, cost-sensitive ordinal classification via regression (COCR), which respects the discrete nature of ordinal ranks in real-world data sets. In particular, COCR applies a theoretically sound method for reducing an ordinal classification to binary and solves the binary classification sub-tasks with point-wise regression. Furthermore, COCR allows us to specify mis-ranking costs to further improve the ranking performance; this ability is exploited by deriving a corresponding cost for a popular ranking criterion, expected reciprocal rank (ERR). The resulting ERR-tuned COCR boosts the benefits of the efficiency of using point-wise regression and the accuracy of top-rank prediction from the ERR criterion. Evaluations on four large-scale benchmark data sets, i.e., "Yahoo! Learning to Rank Challenge" and "Microsoft Learning to Rank,” verify the significant superiority of COCR over commonly used regression approaches.
關聯 Information Retrieval, 17(1), 1-20
資料類型 article
DOI http://dx.doi.org/10.1007/s10791-013-9219-2
dc.contributor 資科系en_US
dc.creator (作者) Ruan, Yu-Xun ; Lin, Hsuan-Tien ; Tsai, Ming-Fengen_US
dc.creator (作者) 蔡銘峰zh_TW
dc.date (日期) 2014.02en_US
dc.date.accessioned 6-Mar-2014 16:29:52 (UTC+8)-
dc.date.available 6-Mar-2014 16:29:52 (UTC+8)-
dc.date.issued (上傳時間) 6-Mar-2014 16:29:52 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/64483-
dc.description.abstract (摘要) This paper proposes a novel ranking approach, cost-sensitive ordinal classification via regression (COCR), which respects the discrete nature of ordinal ranks in real-world data sets. In particular, COCR applies a theoretically sound method for reducing an ordinal classification to binary and solves the binary classification sub-tasks with point-wise regression. Furthermore, COCR allows us to specify mis-ranking costs to further improve the ranking performance; this ability is exploited by deriving a corresponding cost for a popular ranking criterion, expected reciprocal rank (ERR). The resulting ERR-tuned COCR boosts the benefits of the efficiency of using point-wise regression and the accuracy of top-rank prediction from the ERR criterion. Evaluations on four large-scale benchmark data sets, i.e., "Yahoo! Learning to Rank Challenge" and "Microsoft Learning to Rank,” verify the significant superiority of COCR over commonly used regression approaches.en_US
dc.format.extent 488520 bytes-
dc.format.mimetype application/pdf-
dc.language.iso en_US-
dc.relation (關聯) Information Retrieval, 17(1), 1-20en_US
dc.subject (關鍵詞) List-wise ranking ; Cost-sensitive ; Regression ; Reductionen_US
dc.title (題名) Improving Ranking Performance with Cost-sensitive Ordinal Classification via Regressionen_US
dc.type (資料類型) articleen
dc.identifier.doi (DOI) 10.1007/s10791-013-9219-2en_US
dc.doi.uri (DOI) http://dx.doi.org/10.1007/s10791-013-9219-2en_US