摘要: 任务分解被广泛应用于解决大而复杂的问题,学者们已经提出了很多分解模型。知识特征驱动的任务分解模型在无需过多先验知识的情况下,就可以将原始问题分解成一系列子问题,然而这种分解方式却没有考虑对子问题噪点进行处理。在知识特征驱动下,利用马氏距离可以去除子问题的噪点,并对子问题空间进行扩充,这就得到了一种去除噪点的知识特征驱动的任务分解模型。 该模型在处理双螺旋问题、UCI abalone数据集、UCI yeast数据集时,都得到了较高的精度,说明了其可行性和有效性。
[1] Han Jia-wei,Kamber V.数据挖掘:概念与技术 (第2版) [M].2007:277-280 [2] Zhu L,Yuan G,Du Q.An efficient explicit/implicit domain decomposition method for convection-diffusion equations [J].Numerical Methods for Partial Differential Equations,2010,26(4):852-873 [3] Bazán F S V,Gratton S.An Explicit Jordan Decomposition ofCompanion Matrices [J].TEMA Tend.Mat.Apl.Comput,2006,7:209-218 [4] Rajkumar Murthy B E.Parallel alternating explicit implicit domain decomposition algorithm [D].Texas Tech University,2006 [5] Jenkins R E,Yuhas B P.A simplified neural network solutionthrough problem decomposition:The case of the truck backer-upper [J].IEEE Transactions on Neural Networks,1993,4(4):718-720 [6] Thiria S,Me jia C,Badran F,et al.Multimodular architecture forremote sensing operations [C]∥ Moody J E,Hanson S J,Lippmann R P,eds.Advances in Neural Information Processing Systems4.San Mateo,CA:Morgan Kaufmann,1992:675-68 [7] Rifkin R,Klautau A.In defense of one-vs-all classification[J].The Journal of Machine Learning Research,2004,5:101-141 [8] Perdisci R,Gu G,Lee W.Using an ensemble of one-class svm classifiers to harden payload-based anomaly detection systems[C]∥Data Mining,2006.ICDM’06.Sixth International Confe-rence on.IEEE,2006:488-498 [9] Martlnez J,Iglesias C,Matlas J M,et al.DAGSVM Multiclass algorithm based on SVM binary classifiers with 1vsAll approach to the slate tile classification problem[C]∥Modelling for Engineering and Human Behaviour.2012:115 [10] McGill K C,Cummins K L,Dorfman L J.Automatic decomposition of the clinical electromyogram[J].IEEE Transactions on Biomedical Engineering,1985(7):470-477 [11] Cobo L C,Isbell C L Jr,Thomaz A L.Automatic task decomposition and state abstraction from demonstration[C]∥Procee-dings of the 11th International Conference on Autonomous Agents and Multiagent Systems-Volume 1.International Foundation for Autonomous Agents and Multiagent Systems.2012:483-490 [12] Hasan M K,Apu M S,Molla M K I.A robust method for parameter estimation of AR systems using empirical mode decomposition[J].Signal,Image and Video Processing,2010,4(4):451-461 [13] 王国胤,何晓.一种不确定性条件下的自主式知识学习模型[J].软件学报,2003,14(6):1096-1102 [14] 史忠植.知识发现[M].北京:清华大学出版社,2002 [15] Wang G,Wang Y.3DM:domain-oriented data-driven data mi-ning[J].Fundamenta Informaticae,2009,90(4):395-426 [16] 王燕,申元霞,陶春梅.面向领域的数据驱动自主式知识获取模型及实现[J].重庆邮电大学学报:自然科学版,2009,21(4):502-506 [17] Kuntz P,Guillet F,Lehn R,et al.A user-driven process for mining association rules [M]∥Principles of data mining and know-ledge discovery.Berlin Heidelberg:Springer,2000:483-489 [18] Han J,Lakshmanan L V S,Ng R T.Constraint-based,multidimensional data mining [J].Computer,1999,32(8):46-50 [19] Patrick J,Palko D,Munro R,et al.User driven example-based training for creating lexical knowledgebases [C]∥Australasian Natural Language Processing Workshop.Canberra,Australia,2002:17-24 [20] Mahalanobis P C.On the generalized distance in statistics[J].Proceedings of the national institute of sciences of India,1936,2(1):49-55 [21] http://emergent.brynmawr.edu/eprg/?page=TwoSpiralsPro-blem [22] 王国胤,施鸿宝,邓伟.基于NARA 模型和筛选方法的并行神经网络体系结构[J].计算机学报,1996,19(9):679-686 [23] Bender,Timothy,et al.Partitioning Strategies for Modular Neural Networks[C]∥International Joint Conference on Neural Networks (IJCNN 2009).Atlanta GA,2009:296-301 [24] Ikuta C,Uwate Y,Nishio Y.Chaos Glial Network Connected to Multi-Layer Perceptron for Solving Two-Spiral Problem[C]∥Proc.ISCAS’10.May 2010 [25] Clark D,Schreter Z,Adams A.A quantitative comparison ofdystal and back propagation[C]∥Proc.Austr.Conf.Neural Netw(ACNN).1996 [26] Truemper K.Improved comprehensibility and reliability of ex-planations via restricted half space discretization[C]∥Procee-dings of International Conference on Machine Learning and Data Mining(MLDM 2009).2009 [27] Sevilla Villanueva B,Sánchez Marrè M.Providing intelligent decision support systems with flexible data-intensive case-based reasoning[C]∥International Congress on Environmental Mo-delling and Software.2012 |
No related articles found! |
|