Learning to rank with document ranks and scores |
| |
| Authors: | Yan Pan Hai-Xia Luo Yong Tang Chang-Qin Huang |
| |
| Affiliation: | 1. School of Software, Sun Yat-sen University, Guangzhou 510006, China;2. Department of Computer Science, Sun Yat-sen University, Guangzhou 510006, China;3. Department of Computer Science, South China Normal University, Guangzhou 510631, China;4. Engineering Research Center of Computer Network and Information Systems, South China Normal University, Guangzhou 510631, China;1. KU Leuven, Department of Electrical Engineering, ESAT-STADIUS, B-3001, Leuven, Belgium;2. Department of Information Technology, Uppsala University, SE-751 05, Uppsala, Sweden;3. School of Mathematical Sciences, Fudan University, 200433, Shanghai, PR China;1. Department of CS&E, National Institute of Technology, Rourkela-769008, India;2. J E Society K A Lokapur College, Athani-591304, India;1. School of Electrical Engineering, Beijing Jiaotong University, Beijing 100044, China;2. University at Buffalo, The State University of New York, Buffalo, NY 14260, USA;3. Tianjin Key Laboratory of Cognitive Computing and Application, Tianjin University, Tianjin 300072, China;1. ReDCAD Laboratory, University of Sfax, National School of Engineers of Sfax, B.P. 1173, 3038 Sfax, Tunisia;2. Miracl Laboratory, University of Sfax, Pole Technologique de Sfax: Route de Tunis Km 10 B.P. 242, 3021 Sfax, Tunisia |
| |
| Abstract: | The problem of “Learning to rank” is a popular research topic in Information Retrieval (IR) and machine learning communities. Some existing list-wise methods, such as AdaRank, directly use the IR measures as performance functions to quantify how well a ranking function can predict rankings. However, the IR measures only count for the document ranks, but do not consider how well the algorithm predicts the relevance scores of documents. These methods do not make best use of the available prior knowledge and may lead to suboptimal performance. Hence, we conduct research by combining both the document ranks and relevance scores. We propose a novel performance function that encodes the relevance scores. We also define performance functions by combining our proposed one with MAP or NDCG, respectively. The experimental results on the benchmark data collections show that our methods can significantly outperform the state-of-the-art AdaRank baselines. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
