[1]林锦,胡家琛,刘莞玲,等.利用MISA多目标优化的置信规则库分类算法[J].智能系统学报,2019,14(05):982-990.[doi:10.11992/tis.201809022]
 LIN Jin,HU Jiachen,LIU Wanling,et al.Belief rule base classification algorithm using MISA multi-objective optimization[J].CAAI Transactions on Intelligent Systems,2019,14(05):982-990.[doi:10.11992/tis.201809022]
点击复制

利用MISA多目标优化的置信规则库分类算法(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第14卷
期数:
2019年05期
页码:
982-990
栏目:
出版日期:
2019-09-05

文章信息/Info

Title:
Belief rule base classification algorithm using MISA multi-objective optimization
作者:
林锦 胡家琛 刘莞玲 吴英杰
福州大学 数学与计算机科学学院, 福建 福州 350116
Author(s):
LIN Jin HU Jiachen LIU Wanling WU Yingjie
College of Mathematics and Computer Science, Fuzhou University, Fuzhou 350116, China
关键词:
置信规则库分类系统多目标优化多目标免疫系统算法帕累托优化差分进化自适应网格特征属性约减
Keywords:
belief rule base (BRB)classification systemmulti-objective optimizationMISAPareto optimaldifferential evolutionadaptive meshfeature attribute reduction
分类号:
TP18
DOI:
10.11992/tis.201809022
摘要:
现有基于置信规则库的分类系统的分类准确率和效率受到系统参数设置以及规则库结构合理性的影响。为了寻找到最佳的参数值和最优的规则库结构,本文结合多目标免疫系统算法(multiobjective immune system algorithm, MISA)提出利用MISA多目标优化的置信规则库分类算法。该方法融合特征属性约简思想和差分进化算法思想建立训练模型,采用多目标免疫系统算法对系统复杂度和分类准确率进行多目标优化,从而寻找到分类模型的最优解。在实验分析中,首先将本文提出的置信规则库多目标分类系统MISA-BRM和置信规则库分类系统的实验结果进行对比,从复杂度和准确率两个维度说明本文方法的有效性。同时还将本文方法与现有的其他分类方法进行比较,验证本文方法的可行性和有效性。实验结果表明,本文方法能够有效地对基于置信规则库的分类系统的准确率和复杂度进行多目标优化。
Abstract:
The efficiency and accuracy of a belief rule base (BRB) classification system are limited to the determination of the systematic parameters and the structure of the rule base. In this study, we propose the usage of a BRB classification algorithm, i.e., the multi-objective immune system algorithm (MISA), along with multi-objective optimization to determine the optimal parameters and the structure of the rule base. This method simplifies the characteristic attributes using a differential evolution algorithm to develop a training model and subsequently uses MISA to optimize the systematic complexity and the classification accuracy for identifying an optimal solution for the classification model. In the experiment, we initially compare the results of the BRM-based MISA (MISA-BRM) and those of the BRB classification system with respect to their complexity and accuracy to present the benefits of our method. Further, we compare the results with those of the existing classification methods to verify the feasibility and availability of the proposed method. The experimental results denote that the proposed method can effectively optimize the accuracy and complexity of the BRB classification system.

参考文献/References:

[1] 魏茂胜. 数据挖掘中的分类算法综述[J]. 网络安全技术与应用, 2017(6):65-66 WEI Maosheng. Overview of classification algorithms in data mining[J]. Network security technology and application, 2017(6):65-66
[2] TSAKIRIDIS N L, THEOCHARIS J B, ZALIDIS G C. DECO3R:a differential evolution-based algorithm for generating compact fuzzy rule-based classification systems[J]. Knowledge-based systems, 2016, 105:160-174.
[3] PRUSTY M R, JAYANTHI T, CHAKRABORTY J, et al. Performance analysis of fuzzy rule based classification system for transient identification in nuclear power plant[J]. Annals of nuclear energy, 2015, 76:63-74.
[4] LIUBICH L D, KOVALEVSKA L M, LISYANY M I, et al. TGF-β1 expression by glioma C6 cells in vitro[J]. Experimental oncology, 2017, 39(4):258-263.
[5] YANG Jianbo, LIU Jun, WANG Jin, et al. Belief rule-base inference methodology using the evidential reasoning Approach-RIMER[J]. IEEE transactions on systems, man, and cybernetics-part a:systems and humans, 2006, 36(2):266-285.
[6] JIAO Lianmeng, PAN Quan, DEN?UX T, et al. Belief rule-based classification system:Extension of FRBCS in belief functions framework[J]. Information sciences, 2015, 309:26-49.
[7] 刘莞玲, 王韩杰, 傅仰耿, 等. 基于差分进化算法的置信规则库推理的分类方法[J]. 中国科学技术大学学报, 2016, 46(9):764-773 LIU Wanling, WANG Hanjie, FU Yanggeng, et al. Belief rule based inference methodology for classification based on differential evolution algorithm[J]. Journal of University of Science and Technology of China, 2016, 46(9):764-773
[8] MIRJALILI S, SAREMI S, MIRJALILI S M, et al. Multi-objective grey wolf optimizer:A novel algorithm for multi-criterion optimization[J]. Expert systems with applications, 2016, 47:106-119.
[9] HOU Yan, WU Naiqi, ZHOU Mengchu, et al. Pareto-optimization for scheduling of crude oil operations in refinery via genetic algorithm[J]. IEEE transactions on systems, man, and cybernetics:systems, 2017, 47(3):517-530.
[10] 范立云, 周伟, 刘鹏, 等. 高速电磁阀动态响应的多目标优化[J]. 哈尔滨工程大学学报, 2018, 39(1):53-59 FAN Liyun, ZHOU Wei, LIU Peng, et al. Multi objective optimization of dynamic response of high speed solenoid valve[J]. Journal of Harbin Engineering University, 2018, 39(1):53-59
[11] 童晶. 多目标优化的Pareto解的表达与求取[D]. 武汉:武汉科技大学, 2009. TONG Jing. Expressing and obtaining Pareto solutions in multi-objective optimization problem[D]. Wuhan:Wuhan University of Science and Technology, 2009.
[12] COELLO C A, CORTÉS N C. Solving multiobjective optimization problems using an artificial immune system[J]. Genetic programming and evolvable machines, 2005, 6(2):163-190.
[13] KNOWLES J, OATES M, CORNE D. Advanced multi-objective evolutionary algorithms applied to two problems in telecommunications[J]. BT technology journal, 2000, 18(4):51-65.
[14] FALLAHNEZHAD M, MORADI M H, ZAFERANLOUEI S. A hybrid higher order neural classifier for handling classification problems[J]. Expert systems with applications, 2011, 38(1):386-393.
[15] 李继东. 遗传模糊分类系统构建中规则获取和解释性优化的关键技术研究[D]. 昆明:云南大学, 2011. LI Jidong. Research on key techniques of rule acquisition and interpretive optimization in the construction of genetic fuzzy classification system[D]. Kunming:Yunnan University, 2011.
[16] 刘冲. 基于烟花算法的模糊建模方法研究[D]. 郑州:郑州大学, 2016. Liu Chong. Research on fuzzy modeling method based on fireworks algorithm[D]. Zhengzhou:Zhengzhou University, 2016.

备注/Memo

备注/Memo:
收稿日期:2018-09-13。
基金项目:国家自然科学基金项目(71501047,61773123);福建省自然科学基金项目(2015J01248).
作者简介:林锦,女,1995年生,硕士研究生,主要研究方向为智能图像处理、智能系统;胡家琛,男,1995年生,硕士研究生,主要研究方向为图像识别、决策理论方法、数据挖掘与机器学习;刘莞玲,女,1991年生,硕士研究生,主要研究方向为决策理论方法、数据挖掘与机器学习、智能系统。
通讯作者:刘莞玲.E-mail:380509981@qq.com
更新日期/Last Update: 1900-01-01