一种基于共享度模型的改进Rete算法

孙新; 严西敏; 尚煜茗; 欧阳童; 董阔

doi:10.16383/j.aas.2017.c160674

一种基于共享度模型的改进Rete算法

doi: 10.16383/j.aas.2017.c160674

孙新^1, ,,
严西敏^1,,
尚煜茗^1,,
欧阳童^1,,
董阔^1,

1.
北京理工大学计算机学院北京市海量语言信息处理与云计算应用工程技术研究中心北京 100081

基金项目:

国家高技术研究发展计划（863计划） 2015AA015404

详细信息

作者简介:
严西敏北京理工大学硕士研究生.2015年获得北京理工大学计算机学院学士学位.主要研究方向为人工智能, 机器学习. E-mail: yanximin@foxmail.com

尚煜茗北京理工大学硕士研究生.主要研究方向为大数据, 人工智能.E-mail:yumshang@126.com

欧阳童北京理工大学硕士研究生.主要研究方向为机器学习, 人工智能.E-mail:ouyangtong714@163.com

董阔北京理工大学硕士研究生.主要研究方向为智能医疗, 机器学习.E-mail:dongkuo@163.com

通讯作者:
孙新北京理工大学计算机学院副教授.主要研究方向为人工智能, 机器学习, 智能决策.本文通信作者.E-mail: sunxin@bit.edu.cn

计量
- 文章访问数: 1751
- HTML全文浏览量: 544
- PDF下载量: 698
- 被引次数: 0
出版历程
- 收稿日期: 2016-09-19
- 录用日期: 2017-05-06
- 刊出日期: 2017-09-20

An Improved Rete Algorithm Using Shared Degree Model

SUN Xin^{1
, ,},
YAN Xi-Min^1
,,
SHANG Yu-Ming^1
,,
OUYANG Tong^1
,,
DONG Kuo^1
,

1.
Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Application, School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081

Funds:

National High Technology Research and Development Program of China (863 Program) 2015AA015404

More Information

Author Bio:
Master student at Beijing Institute of Technology. He received his bachelor degree in computer science from the School of Computer Science, Beijing Institute of Technology in 2015. His research interest covers artificial intelligence and machine learning

Master student at Beijing Institute of Technology. His research interest covers big data and artificial intelligence

Master student at Beijing Institute of Technology. His research interest covers machine learning and artificial intelligence

Master student at Beijing Institute of Technology. His research interest covers medical intelligence and artificial intelligence

Corresponding author: SUN Xin Associate professor at the School of Computer Science, Beijing Institute of Technology. Her research interest covers artificial intelligence, machine learning, and intelligent decision making. Corresponding author of this paper

摘要

摘要: 专家系统是人工智能领域的重要分支，其中知识表示和知识推理是专家系统的重要组成部分.Rete算法是一种高效的模式匹配算法，能够解决专家系统中推理效率的问题，但是Rete算法在构建Rete网络和推理过程中存在空间和性能方面问题.本文采取有穷自动机理论的思想，阐述了Rete算法中的模式共享度和节点共享度模型，提出了一种Rete网络构建和推理算法来降低Rete网络的复杂度，提升Rete网络推理的速度.最后实验结果表明，本算法能够降低网络复杂度，提升推理速度.
- 专家系统 /
- Rete算法 /
- 有穷自动机 /
- 推理机
Abstract: Expert system is an important brunch of artificial intelligence. Knowledge representation and inference engine is an important part of the expert system. As an efficient pattern-matching algorithm, Rete algorithm can solve the ratiocination efficiency problem in expert system. However, there always exist the problems such as storage and efficiency in building and using Rete network. In this paper, we introduce the idea of finite automata, and the pattern sharing degree and node sharing degree model into Rete algorithm. We propose a Rete network construction and inference algorithm to reduce the complexity of Rete network and improve the speed of Rete network inference. Finally, experimental results show that it can reduce the complexity of the network greatly and raise the inference speed.
- Expert system /
- Rete /
- finite automaton /
- ratiocination machine
注释:

1) 本文责任编委周涛

HTML全文

图 1 Alpha网络节点共享模型图

Fig. 1 The shared model of alpha network node

下载: 全尺寸图片幻灯片

图 2 Beta网络节点共享模型

Fig. 2 The shared model of beta network node

下载: 全尺寸图片幻灯片

图 3 变换顺序后Beta网络节点共享模型

Fig. 3 The shared model of beta network node shu†ed

下载: 全尺寸图片幻灯片

图 4 节点共享度相同示例图

Fig. 4 The situation that the node shared degree is equal

下载: 全尺寸图片幻灯片

表 1 网络构建对比实验结果(ms)

Table 1 The result of network construction (ms)

数据集	序号	Rete		Rete Shuffled		ReteSDM
		构建时间	共享性	构建时间	共享性	构建时间	共享性
Adult	1	38	0.095	43	0.301	53	0.077
	2	31	0.095	38	0.301	47	0.077
	3	31	0.095	38	0.302	55	0.077
	4	32	0.095	37	0.302	47	0.077
	AVG	33	0.095	39	0.301	50.5	0.077
Bank marketing	1	61	0.077	64	0.296	72	0.059
	2	61	0.077	66	0.296	75	0.059
	3	61	0.077	66	0.296	74	0.059
	4	60	0.077	62	0.295	72	0.059
	AVG	60.75	0.077	64.5	0.295	73.25	0.059
Nursery	1	3	0.492	3	0.549	5	0.477
	2	4	0.492	4	0.554	5	0.477
	3	4	0.492	4	0.538	5	0.477
	4	3	0.492	3	0.518	5	0.477
	AVG	3.5	0.492	3.5	0.539	5	0.477

下载: 导出CSV

表 2 推理性能对比实验结果(s)

Table 2 The result of ratiocination effiency (s)

数据集	分组	Rete	Rete Shuffled	ReteSDM	Hashed ReteSDM	偏差率
Adult	1	0.865	0.962	0.736	0.535	0
	2	1.721	1.950	1.472	1.071	0
	3	2.633	2.735	2.202	1.553	0
Bank marketing	1	0.996	1.115	0.823	0.625	0
	2	2.102	2.327	1.650	1.256	0
	3	3.001	3.423	2.436	1.903	0
Nursery	1	0.105	0.115	0.089	0.080	0
	2	0.203	0.217	0.193	0.159	0
	3	0.312	0.332	0.290	0.250	0

下载: 导出CSV

参考文献(18)

[1]	Bemley J L. An introduction to expert systems. In: Proceedings of the 7th Annual National Convention and 10th Anniversary of Black Data Processing Associates "A Decade of Professional Growth". Washington, DC, USA, 1985. 27-34
[2]	Giarratano J C, Riley G D. Expert Systems: Principles and Programming. New York: Thomson Learning, 2005. 78-122
[3]	Munaiseche C P C, Liando O E S. Evaluation of expert system application based on usability aspects. In: Proceedings of the 2016 International Conference on Innovation in Engineering and Vocational Education. Bandung, Indonesia: IOP, 2016. 128-135 http://adsabs.harvard.edu/abs/2016MS%26E..128a2001M
[4]	Ochab M, Wajs W. Expert system supporting an early prediction of the bronchopulmonary dysplasia. Computers in Biology and Medicine, 2016, 69(1): 236-244 https://www.researchgate.net/publication/281467838_Expert_system_supporting_an_early_prediction_of_the_Bronchopulmonary_Dysplasia
[5]	Forgy C L. Rete: a fast algorithm for the many pattern/many object pattern match problem. Artificial Intelligence, 1982, 19(1): 17-37 doi: 10.1016/0004-3702(82)90020-0
[6]	Sipser M. Introduction to the Theory of Computation (3rd edition). Boston: Cengage Learning, 2013. 44-80
[7]	Kornoushenko E K. Operation of closure on the set of states of an incompletely defined deterministic automaton. Cybernetics, 1976, 12(1): 10-17 doi: 10.1007/BF01070337
[8]	Topencharov V, Peeva K. Mathematical automata theory: a categorical approach. Problems of Control and Information Theory, 1980, 9(2): 141-161 http://www.ams.org/mathscinet-getitem?mr=561809
[9]	Ilie L, Yu S. Reducing NFAs by invariant equivalences. Theoretical Computer Science, 2003, 306(1): 373-390 http://www.sciencedirect.com/science/article/pii/S0304397503003116
[10]	Perlin M. Topologically traversing the RETE network. Applied Artificial Intelligence, 1990, 4(3): 155-177 doi: 10.1080/08839519008927948
[11]	Shatueva T A. Method for classical Rete algorithm efficiency improvement for the system of production rules aimed at knowledge extraction from terminological dictionaries. Control Systems and Information Technologies, 2012, 4(3): 31-34
[12]	Xiao D, Zhong X A. Improving Rete algorithm to enhance performance of rule engine systems. In: Proceedings of the 2010 International Conference on Computer Design and Applications. Qinhuangdao, China: IEEE, 2010. 572-575 http://ieeexplore.ieee.org/document/5541368/
[13]	UCI Machine Learning Repository. Adult data set [Online], available: http://archive.ics.uci.edu/ml/datasets/Adult, February 3, 2017
[14]	UCI Machine Learning Repository. Bank marketing data set [Online], available: http://archive.ics.uci.edu/ml/datasets/Bank+Marketing, February 3, 2017
[15]	UCI Machine Learning Repository. Nursery data set [Online], available: http://archive.ics.uci.edu/ml/datasets/Nursery, February 3, 2017
[16]	段艳杰, 吕宜生, 张杰, 赵学亮, 王飞跃.深度学习在控制领域的研究现状与展望.自动化学报, 2016, 42(5): 643-654 http://www.aas.net.cn/CN/abstract/abstract18852.shtml Duan Yan-Jie, Lv Yi-Sheng, Zhang Jie, Zhao Xue-Liang, Wang Fei-Yue. Deep learning for control: the state of the art and prospects. Acta Automatica Sinica, 2016, 42(5): 643-654 http://www.aas.net.cn/CN/abstract/abstract18852.shtml
[17]	俞斌峰, 季海波.稀疏贝叶斯混合专家模型及其在光谱数据标定中的应用.自动化学报, 2016, 42(4): 566-579 http://www.aas.net.cn/CN/abstract/abstract18844.shtml Yu Bin-Feng, Ji Hai-Bo. Sparse bayesian mixture of experts and its application to spectral multivariate calibration. Acta Automatica Sinica, 2016, 42(4): 566-579 http://www.aas.net.cn/CN/abstract/abstract18844.shtml
[18]	管皓, 薛向阳, 安志勇.深度学习在视频目标跟踪中的应用进展与展望.自动化学报, 2016, 42(6): 834-847 http://www.aas.net.cn/CN/abstract/abstract18874.shtml Guan Hao, Xue Xiang-Yang, An Zhi-Yong. Advances on application of deep learning for video object tracking. Acta Automatica Sinica, 2016, 42(6): 834-847 http://www.aas.net.cn/CN/abstract/abstract18874.shtml