[1]李伟宁,王磊.基于ListNet排序学习的特征处理方法[J].计算机技术与发展,2018,28(09):30-33.[doi:10.3969/j.issn.1673-629X.2018.09.007]
　LI Wei-ning,WANG Lei.A Feature Processing Method Based on Ranking Algorithm ListNet[J].,2018,28(09):30-33.[doi:10.3969/j.issn.1673-629X.2018.09.007]
点击复制

基于ListNet排序学习的特征处理方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年09期
页码:
30-33
栏目:
智能、算法、系统工程
出版日期:
2018-09-10

文章信息/Info

Title:
A Feature Processing Method Based on Ranking Algorithm ListNet
文章编号:
1673-629X(2018)09-0030-04
作者:
李伟宁1;  王磊2
1.南京邮电大学 计算机学院,江苏 南京,210003; 2.南京邮电大学 电子科学与工程学院,江苏 南京,210003
Author(s):
LI Wei-ning1 ; WANG Lei2
1. School of Computer,Nanjing University of Posts and Telecommunications,Nanjing 210003,China; 2. School of Electronic Science and Engineering,Nanjing University of Posts and Telecommunications, Nanjing 210003,China
关键词:
信息检索;  排序学习;  特征处理;  ListNet
Keywords:
information retrieval; learning to rank; feature selection; ListNet
分类号:
TP301
DOI:
10.3969/j.issn.1673-629X.2018.09.007
文献标志码:
A
摘要:
排序学习(learning to rank)是一种机器学习与信息检索的交叉学科,可以从大量的包含标记的训练集中自动学习排序模型.特征选取对于排序模型的预测结果有很大的影响,而排序学习对其特征领域的研究却很少.针对这一问题,提出一种特征处理方法:利用基于主成分分析(PCA)的特征重组方法扩展数据集,然后在扩展后的数据集上进行排序算法隐含的特征选择.在LETOR4.0数据集(MQ2007,MQ2008)上基于排序评测函数对ListNet排序算法进行验证.通过对比特征处理前后的排序性能差异,以及添加新特征的个数对排序结果的影响,实验结果表明,经过特征处理的利用排序学习算法构建的排序函数一般要优于原始的排序函数.
Abstract:
Learning to rank is an interdisciplinary of machine learning and information retrieval and learns ranking model automatically from given training data set. The feature space has a great influence on the performance of learning to rank approach,however,there are a little research in terms of feature generation. For this,we propose one feature analysis method which extends data set by feature recom- bination based on PCA,and then performs feature selection implied by learning to rank methods on the extended data set. We evaluate ranking algorithm ListNet on the LETOR4.0 (MQ2007,MQ2008) data set based on ranking evaluation index,and experimentally com- pare the performance of ListNet using the data set with new feature vectors and not,as well as the impact of the number of the new fea- tures added to the result of sort. The experiment shows that ranking functions learned through learning to rank method based on the fea- ture analysis methods outperform the original ones.

相似文献/References:

[1]汪小珍 李龙澍.基于模糊集的信息检索方法[J].计算机技术与发展,2010,(02):37.
　WANG Xiao-zhen,LI Long-shu.An Information Retrieval Scheme Based on Fuzzy Set[J].,2010,(09):37.
[2]杜光芹 张化祥 赵瑞东.主题Web挖掘研究[J].计算机技术与发展,2008,(02):94.
　DU Guang-qin,ZHANG Hua-xiang,ZHAO Rui-dong.State of Topic Web Mining[J].,2008,(09):94.
[3]李桂华 汪学明.语义信息检索框架设计及其算法研究[J].计算机技术与发展,2010,(08):41.
　LI Gui-hua,WANG Xue-ming.Research of Framework and Algorithm of Semantic Information Retrieval[J].,2010,(09):41.
[4]周瑛 张铃.模糊集方法在检索评价系统中的应用[J].计算机技术与发展,2007,(01):111.
　ZHOU Ying,ZHANG Ling.Application of Fuzzy Measure in Information Retrieval Evaluation[J].,2007,(09):111.
[5]张丽坤 蒋波.基于本体的语义Web研究[J].计算机技术与发展,2007,(06):116.
　ZHANG Li-kun,JIANG Bo.Research on Ontology- Based Semantic Web[J].,2007,(09):116.
[6]杨文忠 章兢.用信息-摘要算法提高Web信息检索效率的研究[J].计算机技术与发展,2006,(06):222.
　YANG Wen-zhong,ZHANG Jing.Using Message- Digest Algorithm for improving Efficiency of Web information Searching[J].,2006,(09):222.
[7]王预.数字图书馆信息检索技术及其应用[J].计算机技术与发展,2006,(10):226.
　WANG Yu.Information Retrieval Technique of Digital Library and Its Application[J].,2006,(09):226.
[8]周锦程 王丹 余泉 张维.基于Lucene的全文检索系统的研究与实现[J].计算机技术与发展,2011,(03):67.
　ZHOU Jin-cheng,WANG Dan,YU Quan,et al.Research and Implementation of Full-Text Retrieval Engine Based on Lucene[J].,2011,(09):67.
[9]何拥军 龚发根.基于用户辅助估计的相关网页搜索聚类[J].计算机技术与发展,2011,(07):112.
　HE Yong-jun,GONG Fa-gen.Clustering of Related Pages Based User-Assisted Estimation[J].,2011,(09):112.
[10]黄名选 冯平 谢统义.基于语词抽取与负关联规则挖掘的信息检索[J].计算机技术与发展,2012,(05):157.
　HUANG Ming-xuan,FENG Ping,XIE Tong-yi.Information Retrieval Based on Terms Extraction and Negative Association Rules Mining[J].,2012,(09):157.

更新日期/Last Update: 2018-09-10