山东大学学报 (工学版) ›› 2024, Vol. 54 ›› Issue (3): 36-43.doi: 10.6040/j.issn.1672-3961.0.2023.154
• 机器学习与数据挖掘 • 上一篇
宋辉,张轶哲*,张功萱,孟元
SONG Hui, ZHANG Yizhe*, ZHANG Gongxuan, MENG Yuan
摘要： 针对传统集成学习方法忽略不同样本需使用不同模型权重的问题,提出一种基于类权重和最小化预测熵(class and entropy weights, CEW)的测试时集成方法。类权重为模型预测结果与验证集上各类概率对错分布的相似度,利用欧氏距离计算相识度;在最小化熵过程中,线性组合模型预测经过类权重模块加权后的输出,寻找最小预测熵对应的线性组合作为熵权重,提高集成模型预测能力。试验结果表明:在4个公开医学图像数据集上,CEW方法与最优单一模型相比,平均召回率提高0.23%~2.81%,准确率提高0.5%~2.54%;与DS方法相比,CEW方法平均召回率最多提高1.25%,准确率最多提高1.1%。基于CEW的测试时集成方法能够在测试时(无标签情况下)动态调整模型权重,比同类方法的预测精度更高。
中图分类号:
| [1] DONG X, YU Z, CAO W, et al. A survey on ensemble learning[J]. Frontiers of Computer Science, 2020, 14: 241-258. [2] YANG Y, LV H, CHEN N. A survey on ensemble learning under the era of deep learning[J]. Artificial Intelligence Review, 2023, 56(6): 5545-5589. [3] XIAO Y, WU J, LIN Z, et al. A deep learning-based multi-model ensemble method for cancer prediction[J]. Computer Methods and Programs in Biomedicine, 2018, 153: 1-9. [4] SHARIFANI K, AMINI M. Machine learning and deep learning: a review of methods and applications[J]. World Information Technology and Engineering Journal, 2023, 10(7): 3897-3904. [5] XUE D, ZHOU X, LI C, et al. An application of transfer learning and ensemble learning techniques for cervical histopathology image classification[J]. IEEE Access, 2020, 8: 104603-104618. [6] MIENYE I D, SUN Y. A survey of ensemble learning: concepts, algorithms, applications, and prospects[J]. IEEE Access, 2022, 10: 99129-99149. [7] ZHANG W, LI H, HAN L, et al. Slope stability prediction using ensemble learning techniques: a case study in Yunyang County, Chongqing, China[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2022, 14(4): 1089-1099. [8] QUMMAR S, KHAN F G, SHAH S, et al. A deep learning ensemble approach for diabetic retinopathy detection[J]. IEEE Access, 2019, 7: 150530-150539. [9] KRAWCZYK B, WOZNIAK M. Untrained weighted classifier combination with embedded ensemble pruning[J]. Neurocomputing, 2016, 196: 14-22. [10] HARANGI B. Skin lesion classification with ensembles of deep convolutional neural networks[J]. Journal of Biomedical Informatics, 2018, 86: 25-32. [11] PACHECO A G C, TRAPPENBERG T, KROHLING R A. Learning dynamic weights for an ensemble of deep models applied to medical imaging classification[C] //2020 International Joint Conference on Neural Networks(IJCNN). Glasgow, UK: IEEE, 2020: 1-8. [12] GU R. Multiscale Shannon entropy and its application in the stock market[J]. Physica A: Statistical Mechanics and its Applications, 2017, 484: 215-224. [13] 姜茸, 廖鸿志, 杨明. 信息熵在软件领域中的应用研究现状[J]. 自动化技术与应用, 2015(4): 1-6. JIANG Rong, LIAO Hongzhi, YANG Ming. The current research of information entropy in software domain[J]. Techniques of Automation and Applications, 2015(4): 1-6. [14] ZHOU X, WANG X, HU C, et al. An analysis on the relationship between uncertainty and misclassification rate of classifiers[J]. Information Sciences, 2020, 535: 16-27. [15] WANG D, SHELHAMER E, LIU S, et al. Fully test-time adaptation by entropy minimization[EB/OL].(2020-06-18)[2023-10-16]. https://arxiv.org/abs/2006.10726. [16] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C] //Proceedings of the IEEE conference on computer vision and pattern recognition. Las Vegas, NV, USA: IEEE, 2016: 770-778. [17] HUANG B, LIU J, ZHANG Q, et al. Identification and classification of aluminum scrap grades based on the Resnet18 model[J]. Applied Sciences, 2022, 12(21): 11133. [18] LI B, LIMA D. Facial expression recognition via ResNet-50[J]. International Journal of Cognitive Computing in Engineering, 2021, 2: 57-64. [19] HARJOSEPUTRO Y, YUDA I, DANUKUSUMO K P. MobileNets: efficient convolutional neural network for identification of protected birds[J]. IJASEIT(International Journal on Advanced Science, Engineering and Information Technology), 2020, 10(6): 2290-2296. [20] REZAEE K, MOUSAVIRAD S J, KHOSRAVI M R, et al. An autonomous UAV-assisted distance-aware crowd sensing platform using deep ShuffleNet transfer learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(7): 9404-9413. |
| [1] | 聂秀山,巩蕊,董飞,郭杰,马玉玲. 短视频场景分类方法综述[J]. 山东大学学报 (工学版), 2024, 54(3): 1-11. |
| [2] | 索大翔,李波. 基于Gromov-Wasserstein最优传输的输电线路小目标检测方法[J]. 山东大学学报 (工学版), 2024, 54(3): 22-29. |
| [3] | 高泽文,王建,魏本征. 基于混合偏移轴向自注意力机制的脑胶质瘤分割算法[J]. 山东大学学报 (工学版), 2024, 54(2): 80-89. |
| [4] | 李璐,张志军,范钰敏,王星,袁卫华. 面向冷启动用户的元学习与图转移学习序列推荐[J]. 山东大学学报 (工学版), 2024, 54(2): 69-79. |
| [5] | 陈成,董永权,贾瑞,刘源. 基于交互序列特征相关性的可解释知识追踪[J]. 山东大学学报 (工学版), 2024, 54(1): 100-108. |
| [6] | 李家春,李博文,常建波. 一种高效且轻量的RGB单帧人脸反欺诈模型[J]. 山东大学学报 (工学版), 2023, 53(6): 1-7. |
| [7] | 王旭晴,魏伟波,杨光宇,宋金涛,吕婷,潘振宽. 基于算法展开的图像盲去模糊深度学习网络[J]. 山东大学学报 (工学版), 2023, 53(6): 35-46. |
| [8] | 王碧瑶,韩毅,崔航滨,刘毅超,任铭然,高维勇,陈姝廷,刘嘉巍,崔洋. 基于图像的道路语义分割检测方法[J]. 山东大学学报 (工学版), 2023, 53(5): 37-47. |
| [9] | 周晓昕,廖祝华,刘毅志,赵肄江,方艺洁. 融合历史与当前交通流量的信号控制方法[J]. 山东大学学报 (工学版), 2023, 53(4): 48-55. |
| [10] | 于畅,伍星,邓秋菊. 基于深度学习的多视角螺钉缺失智能检测算法[J]. 山东大学学报 (工学版), 2023, 53(4): 104-112. |
| [11] | 宋佳芮,陈艳平,王凯,黄瑞章,秦永彬. 基于Affix-Attention的命名实体识别语义补充方法[J]. 山东大学学报 (工学版), 2023, 53(2): 70-76. |
| [12] | 李旭涛,杨寒玉,卢业飞,张玮. 基于深度学习的遥感图像道路分割[J]. 山东大学学报 (工学版), 2022, 52(6): 139-145. |
| [13] | 袁钺,王艳丽,刘勘. 基于空洞卷积块架构的命名实体识别模型[J]. 山东大学学报 (工学版), 2022, 52(6): 105-114. |
| [14] | 孟令灿,聂秀山,张雪. 基于遮挡目标去除的公交车拥挤度分类算法[J]. 山东大学学报 (工学版), 2022, 52(4): 83-88. |
| [15] | 杨霄,袭肖明,李维翠,杨璐. 基于层次化双重注意力网络的乳腺多模态图像分类[J]. 山东大学学报 (工学版), 2022, 52(3): 34-41. |
|
||
