A statistics based approach for extracting priority rules from trained neural networks

被引：9

作者：

Zhou, ZH ^{[1
]}

Chen, SF ^{[1
]}

Chen, ZQ ^{[1
]}

机构：

[1] Nanjing Univ, State Key Lab Novel Software Technol, Nanjing, Peoples R China

来源：

IJCNN 2000: PROCEEDINGS OF THE IEEE-INNS-ENNS INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOL III | 2000年

关键词：

D O I：

10.1109/IJCNN.2000.861337

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper, a statistics based approach named STARE that is designed to extract symbolic rules from trained neural networks is proposed. STARE deals with continuous attributes in a unique way so that not only different attributes could be discretized to different number of clusters but also unnecessary discretization could be avoided. STARE introduces statistics to the generation and evaluation of priority rules that have concise appearance. Since it is independent of the network architectures and training algorithms, STARE could be applied to diversified neural classifiers. Experimental results show that rules extracted via STARE are comprehensible, compact and accurate.

引用

页码：401 / 406

页数：6

共 50 条

[41] Extracting rules from neural networks using symbolic algorithms:: Preliminary results
Milaré, CR
de Carvalho, ACPLF
Monard, MC
ICCIMA 2001: FOURTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND MULTIMEDIA APPLICATIONS, PROCEEDINGS, 2001, : 384 - 388
[42] Extracting Interpretable Hierarchical Rules from Deep Neural Networks' Latent Space
Wang, Ya
Paschke, Adrian
RULES AND REASONING, RULEML+RR 2023, 2023, 14244 : 238 - 253
[43] Extracting rules from neural networks by pruning and hidden-unit splitting
Setiono, R
NEURAL COMPUTATION, 1997, 9 (01) : 205 - 225
[44] Extracting comprehensible rules from neural networks via genetic algorithms.
Santos, RT
Nievola, JC
Freitas, AA
2000 IEEE SYMPOSIUM ON COMBINATIONS OF EVOLUTIONARY COMPUTATION AND NEURAL NETWORKS, 2000, : 130 - 139
[45] ERANN: An Algorithm to Extract Symbolic Rules from Trained Artificial Neural Networks
Kamruzzaman, S. M.
Hamid, Md. Abdul
Sarkar, A. M. Jehad
IETE JOURNAL OF RESEARCH, 2012, 58 (02) : 138 - 154
[46] Extracting Social Networks from Language Statistics
Hutchinson, Sterling
Louwerse, Max
DISCOURSE PROCESSES, 2018, 55 (07) : 607 - 618
[47] Extracting symbolic rules from trained support vector machines based on the derivative heuristic information
Zhang, Dexian
Yang, Zhixiao
Fan, Yanfeng
Wang, Ziqiang
FOURTH INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY, VOL 1, PROCEEDINGS, 2007, : 592 - +
[48] A neural networks based approach for fast mining characteristic rules
Rahman, M
Yu, XH
Srinivasan, B
ADVANCED TOPICS IN ARTIFICIAL INTELLIGENCE, 1999, 1747 : 36 - 47
[49] Extracting trees from trained SVM models using a TREPAN based approach
Torres, DED
Rocco, CMS
HIS 2005: 5th International Conference on Hybrid Intelligent Systems, Proceedings, 2005, : 353 - 358
[50] Symbolic knowledge extraction from trained neural networks: A sound approach
Garcez, ASD
Broda, K
Gabbay, DM
ARTIFICIAL INTELLIGENCE, 2001, 125 (1-2) : 155 - 207

← 1 2 3 4 5 →