改进型RBF神经网络的多标签算法研究

doi:10.11896/j.issn.1002-137X.2015.04.065

计算机科学 ›› 2015, Vol. 42 ›› Issue (4): 316-320.doi: 10.11896/j.issn.1002-137X.2015.04.065

• 图形图像与模式识别 • 上一篇

改进型RBF神经网络的多标签算法研究

李书玲,刘蓉,刘红

华中师范大学物理科学与技术学院武汉430079,华中师范大学物理科学与技术学院武汉430079,华中师范大学物理科学与技术学院武汉430079

出版日期:2018-11-14 发布日期:2018-11-14
基金资助:
本文受国家社会科学基金:大众分类中标签间语义关系挖掘研究(12BTQ038)资助

Multi-label Learning for Improved RBF Neural Networks

LI Shu-ling, LIU Rong and LIU Hong

Online:2018-11-14 Published:2018-11-14

摘要/Abstract

摘要： 针对已有的RBF神经网络多标签算法未充分考虑多个样本标签之间的关联性,从而导致泛化性能受到一定影响的问题,研究分析了一种改进型RBF神经网络的多标签算法。该算法首先优化隐含层RBF神经网络基函数中心求取算法——k-均值聚类。采用AP聚类自动寻找k值以获得隐含层节点数目,并构造Huffman树来选取初始聚类中心以防k-均值聚类结果陷入局部最优。然后构造体现标签类之间信息的标签计数向量C,并将其与由优化k-均值聚类得到的聚类中心进行线性叠乘,进而改进RBF神经网络基函数中心,建立RBF神经网络。在公共多标签数据集emotion上的实验表明了该算法能够有效地进行多标签分类。

关键词: 多标签学习,RBF神经网络,k-均值聚类,AP聚类

Abstract: A modified multi-label radial basis function (RBF) neural network algorithm that can fully consider the relationship between numbers of sample labels was presented.This improved algorithm is based on the fact that ignoring the relevance between sample labels may cause potential performance loss.The modified algorithm first optimizes the RBF basis function center calculation algorithm in hidden layer,i.e.k-means clustering.AP clustering is used to automatically find k values to obtain the node number of hidden layer and a Huffman tree is constructed to select the initial cluster centers to prevent the k-means clustering results falling into local optimal.Then a label counting vector C that reflects the correlation between the labels is constructed,and it is linearly multiplied with the clustering centers which are obtained through k-means clustering optimization to optimize the RBF basis function center and establish RBF neural network.Experiments using the public multi-label emotion data sets demonstrate the effectiveness of the proposed algorithm.

Key words: Multi-label learning,RBF neural networks,k-means clustering,AP clustering

李书玲,刘蓉,刘红. 改进型RBF神经网络的多标签算法研究[J]. 计算机科学, 2015, 42(4): 316-320. https://doi.org/10.11896/j.issn.1002-137X.2015.04.065

LI Shu-ling, LIU Rong and LIU Hong. Multi-label Learning for Improved RBF Neural Networks[J]. Computer Science, 2015, 42(4): 316-320. https://doi.org/10.11896/j.issn.1002-137X.2015.04.065

参考文献

[1] Bucak S S,Jin R,Jain A K.Multi-label learning with incomplete class assignments[C]∥Proceedings of the IEEE Computer Socie-ty Conference on Computer Vision and Pattern Recognition.Colorado Springs,CO,2011:2801-2808
[2] 张敏灵,周志华.多标签学习[M].北京:清华大学出版社,2011
[3] 李宇峰,黄圣君,周志华.一种基于正则化的半监督多标记学习方法[J].计算机研究与发展,2012,9(6):1272-1278
[4] 孔祥南,黎铭,姜远,等.一种针对弱标记的直推式多标记分类方法[J].计算机研究与发展,2010,7(8):1392-1399
[5] Sanden C,Zhang J Z.Enhancing multi-label music genre classification through ensemble techniques[C]∥Proceedings of the 34^th international ACM SIGIR conference on Research and Development in Information Retrieval.Beijing,China,2011:705-714
[6] Li Guo-zheng,You Ming-yu,Ge Le,et al.Feature selection forsemi-supervised multi-label learning with application to gene function analysis[C]∥Proceedings of the First ACM International Conference on Bioinformatics and Computational Biology.Niagara Falls,New York,2010:354-357
[7] Gopal C,Yang Y.Multi-label classification with meta-level features[C]∥Proceedings of the 33^rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval.Geneva,Switzerand,2010:315-322
[8] 胡微微.基于语义分析的图像多标签标注算法研究[D].上海:华东理工大学,2013
[9] Crammer K,Singer Y.A new family of online algorithms forcategory ranking[C]∥Proceedings of the 25^th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval.New York,ACM,2002:151-158
[10] Kim Y E,Schmidt E M,Migneco R,et al.Music emotion recognition:A state of the art review[C]∥11^th International Society for Music Information Retrieval Conference.Utrecht,Netherlands,2010:255-266
[11] Barutcuoglu Z,Schapire R E,Troyanskaya O G.Hierarchicalmulti-label prediction of gene function [J].Bioinformatics,2006,2(7):830-836
[12] Zhang Min-ling,Zhou Zhi-hua.A review on multi-label learning algorithms[J].IEEE Transactions on Knowledge and Data Engineering,2014,26(18):1819-1837
[13] Zhang Min-ling,Zhou Zhi-hua.ML-kNN:A lazy learning ap-proach to multi-label learning[J].Pattern Recognition,2007,0(7):2038-2048
[14] 张敏灵.一种新型多标记懒惰学习算法[J].计算机研究与发展,2012,49(11):2271-2282
[15] Zhang Min-ling,Zhang Kun.Multi-label learning by exploitinglabel dependency[C]∥Proceedings of the 16^th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.Washington,DC,USA,2010:999-1007
[16] Zhang Min-ling.ML-RBF:RBF neural networks for multi-label learning[J].Natural Processing Letters,2009,9(2):61-74
[17] 刘晓楠,尹美娟,李明涛,等.面向大规模数据的分层近邻传播聚类算法[J].计算机科学,2014,1(3):185-189
[18] 吴晓蓉.K-均值聚类算法初始中心选取相关问题的研究[D].长沙:湖南大学,2008
[19] Mulan 开源数据库emotion数据集[DB/OL].[2014-3-4].http://source-forge.net/projects/mulan/files/datasets/emotions.rar/download Mulan Open source database emotion Data set
[20] Zhou Zhi-hua,Zhang Min-ling,Huang Sheng-jun,et al.Multi-in-stance multi-label learning[J].Artificial Intelligence,2012,6(1):2291-2320
[21] Trohidis K,Tsoumakas G,Kalliris G,et al.Multi-label classification of music by emotion[J].Journal on Audio,Speech,and Music Processing,Sep.2011:4
[22] Godbole S,Sarawagi S.Discriminative methods for multi-labeled classification[C]∥Proceedings of the 8^th Pacific-Asia Conference on Knowledge Discovery and Data Mining.Heidelberg,Berlin,2004:22-30

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

改进型RBF神经网络的多标签算法研究

Multi-label Learning for Improved RBF Neural Networks

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 0