高级搜索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于双错测度的极限学习机选择性集成方法

夏平凡 倪志伟 朱旭辉 倪丽萍

夏平凡, 倪志伟, 朱旭辉, 倪丽萍. 基于双错测度的极限学习机选择性集成方法[J]. 电子与信息学报, 2020, 42(11): 2756-2764. doi: 10.11999/JEIT190617
引用本文: 夏平凡, 倪志伟, 朱旭辉, 倪丽萍. 基于双错测度的极限学习机选择性集成方法[J]. 电子与信息学报, 2020, 42(11): 2756-2764. doi: 10.11999/JEIT190617
Pingfan XIA, Zhiwei NI, Xuhui ZHU, Liping NI. Selective Ensemble Method of Extreme Learning Machine Based on Double-fault Measure[J]. Journal of Electronics & Information Technology, 2020, 42(11): 2756-2764. doi: 10.11999/JEIT190617
Citation: Pingfan XIA, Zhiwei NI, Xuhui ZHU, Liping NI. Selective Ensemble Method of Extreme Learning Machine Based on Double-fault Measure[J]. Journal of Electronics & Information Technology, 2020, 42(11): 2756-2764. doi: 10.11999/JEIT190617

基于双错测度的极限学习机选择性集成方法

doi: 10.11999/JEIT190617
基金项目: 国家自然科学基金(91546108, 71521001),安徽省自然科学基金 (1908085QG298, 1908085MG232),过程优化与智能决策教育部重点实验室开放课题,中央高校基本科研业务费专项资金(JZ2019HGTA0053, JZ2019HGBZ0128)
详细信息
    作者简介:

    夏平凡:女,1994年生,博士生,研究方向为机器学习、人工智能和集成学习等

    倪志伟:男,1963年生,教授,博士生导师,研究方向为人工智能、机器学习和云计算

    朱旭辉:男,1991年生,讲师,硕士生导师,研究方向为进化计算和机器学习

    倪丽萍:女,1981年生,副教授,硕士生导师,研究方向为分形数据挖掘、人工智能和机器学习

    通讯作者:

    朱旭辉 zhuxuhui@hfut.edu.cn

  • 中图分类号: TP391

Selective Ensemble Method of Extreme Learning Machine Based on Double-fault Measure

Funds: The National Natural Science Foundation of China (91546108, 71521001), The Anhui Provincial Natural Science Foundation (1908085QG298, 1908085MG232), The Open Research Fund Program of Key Laboratory of Process Optimization and Intelligent Decision-making, Ministry of Education, The Fundamental Research Funds for the Central Universities (JZ2019HGTA0053, JZ2019HGBZ0128)
  • 摘要: 极限学习机(ELM)具有学习速度快、易实现和泛化能力强等优点,但单个ELM的分类性能不稳定。集成学习可以有效地提高单个ELM的分类性能,但随着数据规模和基ELM数目的增加,计算复杂度会大幅度增加,消耗大量的计算资源。针对上述问题,该文提出一种基于双错测度的极限学习机选择性集成方法(DFSEE),同时从理论和实验的角度进行了详细分析。首先,运用bootstrap 方法重复抽取训练集,获得多个训练子集,在ELM上进行独立训练,得到多个具有较大差异性的基ELM,构成基ELM池;其次,计算出每个基ELM的双错测度,将基ELM按照双错测度的大小进行升序排序;最后,采用多数投票算法,根据顺序将基ELM逐个累加集成,直至集成精度最优,即获得基ELM最优子集成,并分析了其理论基础。在10个UCI数据集上的实验结果表明,较其他方法使用了更小规模的基ELM,获得了更高的集成精度,同时表明了其有效性和显著性。
  • 图  1  在不同基ELM集成规模下基于成对差异性测度排序集成分类准确率的趋势

    表  1  两个分类器的联合分布

    ${f_i}({x_k}) = {y_k}$${f_i}({x_k}) \ne {y_k}$
    ${f_j}({x_k}) = {y_k}$$a$$b$
    ${f_j}({x_k}) \ne {y_k}$$c$$d$
    下载: 导出CSV

    表  2  UCI数据集

    数据集实例个数属性个数类别
    Heart270132
    Cleveland303135
    Bupa34562
    Wholesale44072
    Diabetes76882
    German1000202
    QSAR1055412
    CMC147393
    Spambase4601572
    Wineq-w4898117
    下载: 导出CSV

    表  3  在不同规模基ELM (100, 200, 300)下的集成分类准确率(%)

    数据集100200300
    DFSEE最高平均最低DFSEE最高平均最低DFSEE最高平均最低
    Heart80.4875.0063.6051.1482.4876.3363.6349.2482.2476.5763.6848.57
    Cleveland57.0155.8748.6638.8657.6156.6248.6838.1157.9156.8248.6837.71
    Bupa75.3770.6760.2148.1876.9571.5160.1547.0977.4072.1860.2346.49
    Wholesale94.0489.5682.7874.1994.7890.2682.7873.4895.1590.5982.7373.04
    Diabetes71.8870.6261.8352.6473.1071.4961.7351.0373.7771.8161.7450.65
    German77.4075.1769.6163.6378.0876.1269.6362.8378.5876.4069.6462.62
    QSAR86.2682.6774.4565.6387.7883.6374.5264.9288.2883.8974.4963.98
    CMC62.9960.4554.2146.5763.4161.0354.2545.9263.8861.3354.2345.46
    Spambase80.7877.5770.1363.3281.5578.1770.1262.7981.7078.4270.1362.53
    Wineq-w51.3850.8046.9744.5251.7351.0346.9444.2151.9051.2046.9444.07
    下载: 导出CSV

    表  4  在不同规模基ELM (100, 200, 300)下DFSEE与Bagging分类准确率对比分析(%)

    数据集100200300
    Bagging本文DFSEEnBagging本文DFSEEnBagging本文DFSEEn
    Heart72.1080.481371.6782.481171.7182.2411
    Cleveland49.2557.01449.2557.61649.2557.916
    Bupa65.6175.371264.1476.951264.9877.4015
    Wholesale86.4494.04886.0094.781086.1195.1511
    Diabetes63.7971.88763.5173.10763.6373.778
    German74.1377.401474.4078.08974.3878.589
    QSAR80.2286.26980.4187.78880.4788.289
    CMC58.2262.99958.4463.411258.4563.8813
    Spambase73.3480.781273.3781.551373.4681.7011
    Wineq-w46.5851.38846.5751.731146.5651.9014
    下载: 导出CSV

    表  5  与其他方法在集成精度(%)和集成规模方面对比分析(基ELM规模200)

    数据集本文DFSEEnAGOBnPOBEnMOAGnEP-FPnSCG-Pn
    Heart82.481174.144977.529674.864374.389575.2438
    Cleveland57.61654.432251.0913250.852549.259556.251
    Bupa76.951269.933772.959969.475965.896676.8948
    Wholesale94.781089.673692.749988.592786.119687.859
    Diabetes73.10766.032668.9910266.275263.738965.3058
    German78.08975.153676.609675.303874.478675.1854
    QSAR87.78883.482484.3710083.943280.438882.0237
    CMC63.411259.634760.9210359.675158.469759.5167
    Spambase81.551376.183279.129776.476776.649376.6658
    Wineq-w51.731150.372349.489348.103448.609650.9846
    下载: 导出CSV

    表  6  与其他方法在运行时间方面的对比分析(s)

    数据集本文DFSEEAGOBPOBEMOAGEP-FPSCG-P
    Heart0.8010.960.790.8718.240.86
    Cleveland0.7722.360.731.102.771.10
    Bupa0.8613.970.850.9541.260.95
    Wholesale1.1617.261.151.2721.971.26
    Diabetes1.2917.951.281.4030.401.39
    German1.7912.581.781.8611.581.86
    QSAR2.2914.012.292.3723.552.37
    CMC2.2524.442.212.6230.852.61
    Spambase8.5443.868.528.80110.468.78
    Wineq-w7.7179.187.588.8548.568.82
    下载: 导出CSV
  • HUANG Guangbin, ZHU Qinyu, and SIEW C K. Extreme learning machine: Theory and applications[J]. Neurocomputing, 2006, 70(1/3): 489–501. doi: 10.1016/j.neucom.2005.12.126
    YANG Yifan, ZHANG Hong, YUAN D, et al. Hierarchical extreme learning machine based image denoising network for visual Internet of Things[J]. Applied Soft Computing, 2019, 74: 747–759. doi: 10.1016/j.asoc.2018.08.046
    吴超, 李雅倩, 张亚茹, 等. 用于表示级特征融合与分类的相关熵融合极限学习机[J]. 电子与信息学报, 2020, 42(2): 386–393. doi: 10.11999/JEIT190186

    WU Chao, LI Yaqian, ZHANG Yaru, et al. Correntropy-based fusion extreme learning machine for representation level feature fusion and classification[J]. Journal of Electronics &Information Technology, 2020, 42(2): 386–393. doi: 10.11999/JEIT190186
    陆慧娟, 安春霖, 马小平, 等. 基于输出不一致测度的极限学习机集成的基因表达数据分类[J]. 计算机学报, 2013, 36(2): 341–348. doi: 10.3724/SP.J.1016.2013.00341

    LU Huijuan, AN Chunlin, MA Xiaoping, et al. Disagreement measure based ensemble of extreme learning machine for gene expression data classification[J]. Chinese Journal of Computers, 2013, 36(2): 341–348. doi: 10.3724/SP.J.1016.2013.00341
    LAN Y, SOH Y C, and HUANG Guangbin. Ensemble of online sequential extreme learning machine[J]. Neurocomputing, 2009, 72(13/15): 3391–3395. doi: 10.1016/j.neucom.2009.02.013
    KSIENIEWICZ P, KRAWCZYK B, and WOŹNIAK M M. Ensemble of Extreme Learning Machines with trained classifier combination and statistical features for hyperspectral data[J]. Neurocomputing, 2018, 271: 28–37. doi: 10.1016/j.neucom.2016.04.076
    李炜, 李全龙, 刘政怡. 基于加权的K近邻线性混合显著性目标检测[J]. 电子与信息学报, 2019, 41(10): 2442–2449. doi: 10.11999/JEIT190093

    LI Wei, LI Quanlong, and LIU Zhengyi. Salient object detection using weighted K-nearest neighbor linear blending[J]. Journal of Electronics &Information Technology, 2019, 41(10): 2442–2449. doi: 10.11999/JEIT190093
    YKHLEF H and BOUCHAFFRA D. An efficient ensemble pruning approach based on simple coalitional games[J]. Information Fusion, 2017, 34: 28–42. doi: 10.1016/j.inffus.2016.06.003
    CAO Jingjing, LI Wenfeng, MA Congcong, et al. Optimizing multi-sensor deployment via ensemble pruning for wearable activity recognition[J]. Information Fusion, 2018, 41: 68–79. doi: 10.1016/j.inffus.2017.08.002
    MARTÍNEZ-MUÑOZ G, HERNÁNDEZ-LOBATO D, and SUÁREZ A. An analysis of ensemble pruning techniques based on ordered aggregation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(2): 245–259. doi: 10.1109/TPAMI.2008.78
    MARTÍNEZ-MUÑOZ G and SUÁREZ A. Pruning in ordered bagging ensembles[C]. The 23rd International Conference on Machine learning, New York, USA, 2006: 609-616. doi: 10.1145/1143844.1143921.
    GUO Li and BOUKIR S. Margin-based ordered aggregation for ensemble pruning[J]. Pattern Recognition Letters, 2013, 34(6): 603–609. doi: 10.1016/j.patrec.2013.01.003
    DAI Qun, ZHANG Ting, and LIU Ningzhong. A new reverse reduce-error ensemble pruning algorithm[J]. Applied Soft Computing, 2015, 28: 237–249. doi: 10.1016/j.asoc.2014.10.045
    ZHOU Zhihua, WU Jianxin, and TANG Wei. Ensembling neural networks: Many could be better than all[J]. Artificial Intelligence, 2002, 137(1/2): 239–263. doi: 10.1016/S0004-3702(02)00190-X
    CAVALCANTI G D C, OLIVEIRA L S, MOURA T J M, et al. Combining diversity measures for ensemble pruning[J]. Pattern Recognition Letters, 2016, 74: 38–45. doi: 10.1016/j.patrec.2016.01.029
    MAO Shasha, CHEN Jiawei, JIAO Licheng, et al. Maximizing diversity by transformed ensemble learning[J]. Applied Soft Computing, 2019, 82: 105580. doi: 10.1016/j.asoc.2019.105580
    TANG E K, SUGANTHAN P N, and YAO Xin. An analysis of diversity measures[J]. Machine Learning, 2006, 65(1): 247–271. doi: 10.1007/s10994-006-9449-2
    GIACINTO G and ROLI F. Design of effective neural network ensembles for image classification purposes[J]. Image and Vision Computing, 2001, 19(9/10): 699–707. doi: 10.1016/S0262-8856(01)00045-2
    FUSHIKI T. Estimation of prediction error by using K-fold cross-validation[J]. Statistics and Computing, 2011, 21(2): 137–146. doi: 10.1007/s11222-009-9153-8
    ZHOU Hongfa, ZHAO Xuehan, and WANG Xiao. An effective ensemble pruning algorithm based on frequent patterns[J]. Knowledge-Based Systems, 2014, 56: 79–85. doi: 10.1016/j.knosys.2013.10.024
  • 加载中
图(1) / 表(6)
计量
  • 文章访问数:  2036
  • HTML全文浏览量:  527
  • PDF下载量:  50
  • 被引次数: 0
出版历程
  • 收稿日期:  2019-08-12
  • 修回日期:  2020-06-21
  • 网络出版日期:  2020-07-17
  • 刊出日期:  2020-11-16

目录

    /

    返回文章
    返回