Page 272 - 《软件学报》2025年第10期

丁世飞 等: 面向二分类问题的直觉模糊深度随机配置网络 4669

ecoli3 wpbc 表 4 加入噪声的 Circle-in-the-Square 问题分类精度

0.78
噪声比
0.93 0.77 例 (%) IFTWSVM RVFL SCN IFSCN DSCN IFDSCN
0.76 0 0.994 4 0.978 8 0.994 5 0.996 0 0.996 3 0.998 0
0.92 0.75 5 0.984 5 0.964 0 0.984 1 0.991 2 0.991 2 0.992 3
ACC ACC 0.74 10 0.984 0 0.958 5 0.983 2 0.990 5 0.990 8 0.992 1
0.91
0.73
0.72
0.90
0.71
0 5 10 0 5 10
噪声比例 (%) 噪声比例 (%)
SCN DSCN IFSCN IFDSCN
图 3 不同噪声比例下的实验结果

4 结束语
随机配置网络 (SCN) 能够利用监督机制自适应分配隐含层节点参数．本文引入直觉模糊概念, 提出的
IFDSCN 能够通过构造基于隶属度和非隶属度函数的直觉模糊函数对数据样本进行加权, 并设计了一种新的监督
机制分配隐含层节点参数. 通过复在二分类问题数据集上的实验结果表明, IFDSCN 提高了 SCN、DSCN 的二分
类精度, 且具有更高的鲁棒性.
多分类问题作为分类问题的重要分支, 在未来的研究中, IFDSCN 将用于解决多分类问题, 以获得更高的分类
精度. 除此之外, 受到一些安全学习框架研究的启发 [20,21] , 未来将进行进一步研究, 以提高 IFDSCN 数据预测的安
全性.

References:
[1] Pao YH, Takefuji Y. Functional-link net computing: Theory, system architecture, and functionalities. Computer, 1992, 25(5): 76–79. [doi:
10.1109/2.144401]
[2] Wang DH, Li M. Stochastic configuration networks: Fundamentals and algorithms. IEEE Trans. on Cybernetics, 2017, 47(10):
3466–3479. [doi: 10.1109/TCYB.2017.2734043]
[3] Huang CQ, Huang QH, Wang DH. Stochastic configuration networks based adaptive storage replica management for power big data
processing. IEEE Trans. on Industrial Informatics, 2020, 16(1): 373–383. [doi: 10.1109/TII.2019.2919268]
[4] Wang QJ, Dai W, Ma XP, Shang ZG. Driving amount based stochastic configuration network for industrial process modeling.
Neurocomputing, 2020, 394: 61–69. [doi: 10.1016/j.neucom.2020.02.029]
[5] Wang W, Jia Y, Yu W, Pang HS, Cai KW. On-line ammonia nitrogen measurement using generalized additive model and stochastic
configuration networks. Measurement, 2021, 170: 108743. [doi: 10.1016/j.measurement.2020.108743]
[6] Liu JN, Hao RJ, Zhang TL, Wang XZ. Vibration fault diagnosis based on stochastic configuration neural networks. Neurocomputing,
2021, 434: 98–125. [doi: 10.1016/j.neucom.2020.12.080]
[7] Li M, Wang DH. 2-D stochastic configuration networks for image data analytics. IEEE Trans. on Cybernetics, 2021, 51(1): 359–372.
[doi: 10.1109/TCYB.2019.2925883]
[8] Dai W, Li DP, Zhou P, Chai TY. Stochastic configuration networks with block increments for data modeling in process industries.
Information Sciences, 2019, 484: 367–386. [doi: 10.1016/j.ins.2019.01.062]
[9] Wang DH, Li M. Robust stochastic configuration networks with kernel density estimation for uncertain data regression. Information
Sciences, 2017, 412–413: 210–222. [doi: 10.1016/j.ins.2017.05.047]
[10] Li M, Huang CQ, Wang DH. Robust stochastic configuration networks with maximum correntropy criterion for uncertain data regression.
Information Sciences, 2019, 473: 73–86. [doi: 10.1016/j.ins.2018.09.026]
[11] Wang DH, Cui CH. Stochastic configuration networks ensemble with heterogeneous features for large-scale data analytics. Information
Sciences, 2017, 417: 55–71. [doi: 10.1016/j.ins.2017.07.003]
[12] Pratama M, Wang DH. Deep stacked stochastic configuration networks for lifelong learning of non-stationary data streams. Information
Sciences, 2019, 495: 150–174. [doi: 10.1016/j.ins.2019.04.055]