Listwise approaches based on feature ranking discovery

Listwise approaches are an important class of learning to rank, which utilizes automatic learning techniques to discover useful information. Most previous research on listwise approaches has focused on optimizing ranking models using weights and has used imprecisely labeled training data; optimizing...

Full description

Saved in:
Bibliographic Details
Published inFrontiers of Computer Science Vol. 6; no. 6; pp. 647 - 659
Main Authors WANG, Yongqing, MAO, Wenji, ZENG, Daniel, XIA, Fen
Format Journal Article
LanguageEnglish
Published Heidelberg Higher Education Press 01.12.2012
SP Higher Education Press
Springer Nature B.V
Subjects
Algorithms
Computer Science
feature's ranking discovery
KNN
Learning
learning to rank
listwise approach
Ranking
Research Article
Training
使用权
加法法则
基础
学习技术
性能改进
排除方法
训练数据
listwise approach
learning to rank
feature’s ranking discovery
Online AccessGet full text

Cover

More Information
Summary:Listwise approaches are an important class of learning to rank, which utilizes automatic learning techniques to discover useful information. Most previous research on listwise approaches has focused on optimizing ranking models using weights and has used imprecisely labeled training data; optimizing ranking models using features was largely ignored thus the continuous performance improvement of these approaches was hindered. To address the limitations of previous listwise work, we propose a quasi-KNN model to discover the ranking of features and employ rank addition rule to calculate the weight of combination. On the basis of this, we propose three listwise algorithms, FeatureRank, BLFeatureRank, and DiffRank. The experimental results show that our proposed algorithms can be applied to a strict ordered ranking training set and gain better performance than state-of-the-art listwise algorithms.
Bibliography:11-5731/TP
learning to rank, listwise approach, feature＇s ranking discovery
Listwise approaches are an important class of learning to rank, which utilizes automatic learning techniques to discover useful information. Most previous research on listwise approaches has focused on optimizing ranking models using weights and has used imprecisely labeled training data; optimizing ranking models using features was largely ignored thus the continuous performance improvement of these approaches was hindered. To address the limitations of previous listwise work, we propose a quasi-KNN model to discover the ranking of features and employ rank addition rule to calculate the weight of combination. On the basis of this, we propose three listwise algorithms, FeatureRank, BL-FeatureRank, and DiffRank. The experimental results show that our proposed algorithms can be applied to a strict ordered ranking training set and gain better performance than state-of-the-art listwise algorithms.
Document received on :2011-08-31
Document accepted on :2012-01-10
feature's ranking discovery
listwise approach
learning to rank
ISSN:1673-7350
2095-2228
1673-7466
2095-2236
DOI:10.1007/s11704-012-1170-7