Multi-label Learning of Kernel Extreme Learning Machine with Non-Equilibrium Label Completion
CHENG Yu-sheng1,2, ZHAO Da-wei1, WANG Yi-bin1,2, PEI Gen-sheng1
1. School of Computer and Information, Anqing Normal University, Anqing, Anhui 246011, China;
2. The University Key Laboratory of Intelligent Perception and Computing of Anhui Province, Anqing, Anhui 246011, China
Abstract：At present,many researchers usually directly add the label confidence matrix as a priori knowledge to the classification model,and do not consider the influence of non-equilibrium prior knowledge on the quality of the label set.Based on this,the method of non-equilibrium parameters is introduced,and the basis confidence matrix obtained from the prior knowledge is non-equilibrium.Therefore,a multi-label learning algorithm is proposed,which uses kernel extreme learning machine with non-equilibrium label completion (KELM-NeLC).Firstly,information entropy is used to measure the correlation between labels which gets the basic label confidence matrix.Secondly,the basic label confidence matrix is improved to construct non-equilibrium label completion matrix by the non-equilibrium parameter.Finally,in order to learn to obtain a more accurate label confidence matrix,the non-equilibrium label completion matrix is introduced with the kernel extreme learning machine to solve the multi-label classification problem.The experimental results of the proposed algorithm in the opening benchmark multi-label datasets show that the KELM-NeLC algorithm has some advantages over other comparative multi-label learning algorithms and the statistical hypothesis test further illustrates the effectiveness of the proposed algorithm.
[1] Zhang M L,Zhou Z H.A review on multi-Label learning algorithms[J].IEEE Transactions on Knowledge and Data Engineering,2014,26(8):1819-1837.
[2] Zhang M L,Zhou Z H.ML-KNN:A lazy learning approach to multi-label learning[J].Pattern Recognition,2007,40(7):2038-2048.
[3] Elisseeff A,Weston J.A kernel method for multi-labelled classification[A].Advances in Neural Information Processing Systems 14[C].Cambridge,UK:MIT Press,2002.681-687.
[4] Zhang M L,Wu L.Lift:Multi-label learning with label-specific features[J].Pattern Analysis & Machine Intelligence IEEE Transactions on,2015,37(1):107-120.
[5] Shannon C E.A mathematical theory of communication[J].Bell System Technical Journal,1948,27(3):379-423.
[6] Lee J,Kim H,Kim N R,et al.An approach for multi-label classification by directed acyclic graph with label correlation maximization[J].Information Sciences,2016,351(C):101-114.
[7] Huang G B,Zhu Q Y,Siew C K.Extreme learning machine:theory and applications[J].Neurocomputing,2006,70(1-3):489-501.
[8] Huang G B.An Insight into extreme learning machines:random neurons,random features and kernels[J].Cognitive Computation,2014,6(3):376-390.
[9] Ding S,Xu X,Nie R.Extreme learning machine and its applications[J].Neural Computing & Applications,2014,25(3-4):549-556.
[10] Huang G,Huang G B,Song S,et al.Trends in extreme learning machines:a review[J].Neural Networks the Official Journal of the International Neural Network Society,2015,61(C):32-48.
[11] Luo F,Guo W,Yu Y,et al.A multi-label classification algorithm based on kernel extreme learning machine[J].Neurocomputing,2017,260:313-320.
[12] Liu Y,Wen K,Gao Q,et al.SVM based multi-label learning with missing labels for image annotation[J].Pattern Recognition,2018,78:307-317.
[13] Pizzuti C.A Multi-objective genetic algorithm for community detection in networks[A].IEEE International Conference on TOOLS with Artificial Intelligence[C].Newark,New Jersey,USA:IEEE Computer Society,2009.379-386.
[14] Zhang M L,Zhou Z H.A Review on multi-label learning algorithms[J].IEEE Transactions on Knowledge & Data Engineering,2014,26(8):1819-1837.
[15] 邓万宇,郑庆华,陈琳,等.神经网络极速学习方法研究[J].计算机学报,2010,33(2):279-287.Deng W Y,Zheng Q H,Chen L,et al.Research onextreme learning of neural networks[J].Chinese Journal of Computers,2010,33(2):279-287.(in Chinese)
[16] Zhang Q W,Zhong Y,Zhang M L.Feature-induced labeling information enrichment for multi-label learning[A].Thirty-Second AAAI Conference on Artificial Intelligence[C].New Orleans:AAAI in Press,2018.
[17] 何志芬,杨明,刘会东.多标记分类和标记相关性的联合学习[J].软件学报,2014(9):1967-1981.He Z F,Yang M,Liu H D.Joint learning of multi-label classification and label correlations[J].Journal of Software,2014,25(9):1967-1981.(in Chinese)
[18] Ar J.Statistical comparisons of classifiers over multiple data sets[J].Journal of Machine Learning Research,2006,7(1):1-30.
2019, Vol. 47 
