[1] Liu Huafu, Pan Yi, Wang Zhong,. New rank learning algorithm [J]. Journal of Southeast University (English Edition), 2007, 23 (3): 447-450. [doi:10.3969/j.issn.1003-7985.2007.03.030]

Copy

New rank learning algorithm()

一种新的排序学习算法

Share：

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:

23

Issue:

2007 3

Page:

447-450

Research Field:

Automation

Publishing date:

2007-09-30

Info

Title:

New rank learning algorithm

一种新的排序学习算法

Author(s):

Liu Huafu;  Pan Yi;  Wang Zhong

Department of Computer Science and Technology, Changsha University, Changsha 410003, China

刘华富;  潘怡;  王仲

长沙大学计算机科学与技术系, 长沙 410003

Keywords:

machine learning;  rank learning algorithm;  decision tree;  splitting rule

机器学习;  排序学习算法;  决策树;  分裂规则

PACS:

TP181

DOI:

10.3969/j.issn.1003-7985.2007.03.030

Abstract:

To overcome the limitation that complex data types with noun attributes cannot be processed by rank learning algorithms, a new rank learning algorithm is designed.In the learning algorithm based on the decision tree, the splitting rule of the decision tree is revised with a new definition of rank impurity. A new rank learning algorithm, which can be intuitively explained, is obtained and its theoretical basis is provided.The experimental results show that in the aspect of average rank loss, the ranking tree algorithm outperforms perception ranking and ordinal regression algorithms and it also has a faster convergence speed.The rank learning algorithm based on the decision tree is able to process categorical data and select relative features.

为了克服排序学习算法不能处理包括名词性特征的复杂数据类型的局限性, 设计一种新的排序学习算法.在决策树学习算法中, 采用新的等级不纯度定义, 修改决策树的分裂规则, 得到具有直观解释的排序算法, 并给出了相关理论基础.实验结果表明:排序树的平均等级损失明显优于感知机类算法和序回归类算法, 且具有较快的收敛速度.基于决策树的排序学习算法, 可以处理名词性数据和选择相关的特征.

References:

[1] Cohen W W, Schapire R E, Singer Y.Learning to order things[J].Journal of Artificial Intelligence Research, 1999, 10(5):243-275.
[2] Edward F H.Online ranking/collaborative filtering using the perception algorithm[C]//Proceedings of the 20th International Conference on Machine Learning (ICML-2003).Washington DC, 2003:115-132.
[3] Shen Libin, Aravind K J.Ranking and reranking with perception[J].Machine Learning, 2005, 60(3):73-96.
[4] Freund Y, Iyer R, Schapire R E, et al.An efficient boosting algorithm for combining preference[J].Journal of Machine Learning Research, 2003, 3(4):933-969.
[5] Quinlan R.Induction of decision trees[J].Machine Learning, 1986, 18(1):81-106.
[6] Shaw W M, Wood J B, Wood, R.E, et al.The cystic fibrosis database:content and research opportunities[J].Library and Information Science Research, 1991, 13(5):347-367.
[7] GroupLens Research Project.MovieLens data sets[EB/OL].(2005-09-18)[2007-04-20].http://www.grouplens.org/data.

Memo

Memo:

Biography: Liu Huafu(1961—), male, associate professor, hfliu9063@163.com.

Common functions