|本期目录/Table of Contents|

[1]王晓栋*,严 菲,洪朝群.一种基于半监督多任务学习的特征选择模型[J].厦门大学学报(自然科学版),2017,56(04):567-575.[doi:10.6043/j.issn.0438-0479.201611021]
 WANG Xiaodong*,YAN Fei,HONG Chaoqun.A Feature Selection Framework Based on Semi-supervised Multi-task Learning[J].Journal of Xiamen University(Natural Science),2017,56(04):567-575.[doi:10.6043/j.issn.0438-0479.201611021]
点击复制

一种基于半监督多任务学习的特征选择模型(PDF/HTML)
分享到:

《厦门大学学报(自然科学版)》[ISSN:0438-0479/CN:35-1070/N]

卷:
56卷
期数:
2017年04期
页码:
567-575
栏目:
研究论文
出版日期:
2017-07-26

文章信息/Info

Title:
A Feature Selection Framework Based on Semi-supervised Multi-task Learning
文章编号:
0438-0479(2017)04-0567-09
作者:
王晓栋*严 菲洪朝群
厦门理工学院计算机与信息工程学院,福建 厦门 361024
Author(s):
WANG Xiaodong*YAN FeiHONG Chaoqun
College of Computer and Information Engineering,Xiamen University of Technology,Xiamen 361024,China
关键词:
特征选择 多任务学习 网页自动分类 l21范数
Keywords:
feature selection multi-task learning web page classification l21-norm
分类号:
TP 391.9
DOI:
10.6043/j.issn.0438-0479.201611021
文献标志码:
A
摘要:
为了综合利用流形学习、多任务学习和正则化约束的优势,提出一种基于全局和局部约束的半监督多任务特征选择(semi-supervised multi-task feature selection,SMFS)模型,在多个任务间共享学习的基础上,构建SMFS模型.该模型采用l2,1范数约束选择最具判别性的特征,避免噪声的干扰,并引入局部信息约束提高特征选择的准确度.将SMFS模型应用于网页自动分类,与目前流行的几种算法进行对比,证明了该算法的有效性.
Abstract:
Feature selection,which aims to reduce the dimension of the data and remove the redundant feature,plays an important role in improving the performance of multimedia processing.In this paper,a semi-supervised multi-task feature selection algorithm built on sharing information between multiple learning tasks has been proposed.In order to select the most discriminative features,and avoid the noise interference,we have also constructed a semi-supervised multi-task feature selection model with l2,1-norm and local information constraint.In order to verify the effectiveness of our algorithm,we apply the algorithm to the web page classification application and compare it with several state-of-the-art algorithms.Results show that the proposed algorithm is effective.

参考文献/References:

[1] 李长虹,李堂秋.一种改进的特征选择方法在文本分类系统中的应用[J].厦门大学学报(自然科学版),2005,44(增1):239-242.
[2] WANG H,NIE F,HENG H.Large-scale cross-language web page classification via dual knowledge transfer using fast nonnegative matrix trifactorization[J].ACM Transaction on Knowledge Discovery from Data,2015,10(1):1-29.
[3] 郑艳红,张东站.基于同义词词林的文本特征选择方法[J].厦门大学学报(自然科学版),2012,51(2):200-203.
[4] 曾华琳,周昌乐,陈毅东,等.基于特征自动选择方法的汉语隐喻计算[J].厦门大学学报(自然科学版),2016,55(3):406-412.
[5] ZHAO Z,LIU H.Semi-supervised feature selection via spectral analysis[C]∥Proceedings of the 7th SIAM International Conference on Data Mining.Minneapolis:SIAM,2007:641-646.
[6] MA Z,NIE F,YANG Y,et al.Discriminating joint feature analysis for multimedia data understanding[J].IEEE Transactions on Multimedia,2012,14(6):1662-1672.
[7] NIE F,XU D,TSANG W H,et al.Flexible manifold embedding:a framework for semi-supervised and unsupervised dimension reduction[J].IEEE Transactions on Image Processing,2010,19(7):1921-1932.
[8] CHANG X,NIE F,YANG Y,et al.A convex formulation for semi-supervised multi-label feature selection[C]∥Twenty-Eighth AAAI Conference on Artificial Intelligence.Quebec:AAAI Press,2014:1171-1177.
[9] YANG Y,MA Z,HAUPTMANN A G,et al.Feature selection for multimedia analysis by sharing information among multiple tasks[J].IEEE Transactions on Multimedia,2013,15(3):661-669.
[10] MA Z,YANG Y,CAI Y,et al.Knowledge adaptation for ad hoc multimedia event detection with few exemplars[C]∥Proceedings of the 20th ACM International Conference on Multimedia.New York:ACM,2012:469-478.
[11] MA Z,YANG Y,SEBE N,et al.Knowledge adaptation with partially shared features for event detection with few exemplars[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2014,36(9):1789-1802.
[12] LI Z C,YANG Y,LIU J,et al.Unsupervised feature selection using nonnegative spectral analysis[C]∥Proceedings of 26th AAAI Conference on Artificial Intelligence.Toronto:AAAI Press,2012:1026-1032.
[13] WANG F,WANG X,LI T.Semi-supervised multi-task learning with task regularizations[C]∥2009 Ninth IEEE International Conference on Data Mining.Miami Beach:IEEE,2009:562-568.
[14] QI Y,TASTAN O,CARBONELL J G,et al.Semi-supervised multi-task learning for predicting interactions between HIV-1 and human proteins[J].Bioinformatics.2010,26(18):i645-i652.
[15] CHANG X,YANG Y.Semi-supervised feature analysis by mining correlations among multiple tasks[J].IEEE Transactions on Neural Networks and Learning Systems,2016(99):1-12.
[16] YANG Y,SHEN H T,MA Z G,ea al.l2,1-norm regularized discriminative feature selection for unsupervised learning[C]∥Proceeding of the 22th International Joint Conference on Artificial Intelligence.Barcelona:AAAI Press,2011:1589-1594.
[17] NIE F,CAI X,HUANG H,et al.Efficient and robust feature selection via joint l2,1-norms minimization[C]∥Proceedings of the 23th International Conference on Neural Information Processing Systems.New York:Curran Associates Inc,2011:1813-1821.
[18] 史彩娟,阮秋琦.基于增强稀疏性特征选择的网络图像标注[J].软件学报,2015(7):1800-1811.
[19] GIRALDO FORERO A F,JARAMILLO GARZON J A,RUIZ MUNOZ J F,et al.Managing imbalanced data sets in multi-label problems:a case study with the SMOTE algorithm[C]∥Proceedings of Progress in Pattern Recognition,Image Analysis,Computer Vision,and Applications:18th Iberoamerican Congress.Berlin:Springer,2013:334-342.
[20] ZHANG M,HOU Z.ML-KNN:a lazy learning approach to multi-label learning[J].Pattern Recognition,2007,40(7):2038-2048.

备注/Memo

备注/Memo:
收稿日期:2016-11-06 录用日期:2017-04-04
基金项目:国家自然科学基金(61502405); 福建省自然科学基金(2016J01324,2017J01511)
*通信作者:xdwangjsj@xmut.edu.cn
更新日期/Last Update: 1900-01-01