LOTMAP: learning to optimize top-N recommendation with mean average precision

CASIA OpenIR > 中科院工业视觉智能装备工程实验室 > 精密感知与控制

	LOTMAP: learning to optimize top-N recommendation with mean average precision
	Yang Yehui1 ; Zhang Wensheng1 ; Xie Yuan1 ; Zhu Limin1 ; Tan Yuanhua 2
发表期刊	Innovative Computing, Information and Control Express Lettersv
	2014-12
卷号	8 期号:12 页码:3299-3306
摘要	People tend to pay attention to the top few items given by recommendation system, it's crucial to ensure the precision and personalization at the top of recommendation list. In this paper, we propose a ranking-oriented Collaborative Filtering (CF) algorithm LOTMAP, which aims to optimize the top-N recommendation by directly maximizing Mean Average Precision (MAP). As the relevance between users and items are not clarified in rating datasets, we also introduce a new method that judges the relevance through the explicit rating scores. Experiments on real world datasets show that our algorithm is effective, and outperforms the state-of-the-art CF baselines.
关键词	Top-n Recommendation Collaborative Filtering Optimize Matrix Factorization Mean Average Precision
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/11502
专题	中科院工业视觉智能装备工程实验室_精密感知与控制
作者单位	1.Institute of Automation, Chinese Academy of Sciences 2.$Karamay Hongyou Software CO., LTD
第一作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Yang Yehui,Zhang Wensheng,Xie Yuan,et al. LOTMAP: learning to optimize top-N recommendation with mean average precision[J]. Innovative Computing, Information and Control Express Lettersv,2014,8(12):3299-3306.
APA	Yang Yehui,Zhang Wensheng,Xie Yuan,Zhu Limin,&Tan Yuanhua.(2014).LOTMAP: learning to optimize top-N recommendation with mean average precision.Innovative Computing, Information and Control Express Lettersv,8(12),3299-3306.
MLA	Yang Yehui,et al."LOTMAP: learning to optimize top-N recommendation with mean average precision".Innovative Computing, Information and Control Express Lettersv 8.12(2014):3299-3306.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
ICICEL-Vol8No12.pdf（716KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载