[1]曾子林,陈建军.基于决策树桩的元特征提取[J].江西师范大学学报(自然科学版),2018,(06):616-620.[doi:10.16357/j.cnki.issn1000-5862.2018.06.12]
 ZENG Zilin,CHEN Jianjun.The Extraction of Meta-Feature Based on Decision Stump[J].,2018,(06):616-620.[doi:10.16357/j.cnki.issn1000-5862.2018.06.12]
点击复制

基于决策树桩的元特征提取()
分享到:

《江西师范大学学报》(自然科学版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2018年06期
页码:
616-620
栏目:
信息科学与技术
出版日期:
2018-12-20

文章信息/Info

Title:
The Extraction of Meta-Feature Based on Decision Stump
文章编号:
1000-5862(2018)06-0616-05
作者:
曾子林1陈建军2
1.解放军陆军步兵学院,江西 南昌 330103; 2.上饶职业技术学院,江西 上饶 334109
Author(s):
ZENG Zilin1CHEN Jianjun2
1.Army Infantry College of People's Liberation Army,Nanchang Jiangxi 330103,China; 2.Shangrao Vocational and Technical College,Shangrao Jiangxi 334109,China
关键词:
元特征 算法性能 算法排序 决策树桩
Keywords:
meta-feature performance of algorithms ranking of algorithms decision stump
分类号:
TP 391
DOI:
10.16357/j.cnki.issn1000-5862.2018.06.12
文献标志码:
A
摘要:
“No Free Lunch”定理表明:若无任何先验假设,则没有理由认为一种算法优于另一种算法.算法的性能与问题的元特征密切相关.目前的元特征提取方法只关注从数据集中提取元特征,而忽略了候选算法元特征的提取.为此,在原有元特征集合的基础上提出基于决策树桩的元特征提取方法,将候选算法信息纳入新的元特征集合中.实验表明:在传统元特征集合中加入基于决策树桩的元特征后,算法排序的预测准确率能够得到显著提高.
Abstract:
The "No Free Lunch" theorem shows that there is no reason to think that one algorithm is superior to the other one without any prior assumptions.The performance of algorithm is closely related to the meta-feature of problem.The current meta-feature extraction method is only concerned with extracting meta-feature from the data set,while ignoring the meta-feature extraction of candidate algorithms.Therefore,an extraction method based on decision stump is proposed,which can effectively reflect the information of candidate algorithms.Experiments show that the new meta-feature sets significantly increase the prediction accuracy of algorithm ranking.

参考文献/References:

[1] Wolpert D H,Macready W G.No free lunch theorems for search[J].IEEE Transactions on Evolutionary Computation,1997,1(1):67-82.
[2] Rendell L,Cho H.Empirical learning as a function of concept character[J].Machine Learning,1990,5(3):267-298.
[3] Cruz R M O,Sabourin R,Cavalcanti G D C.Meta-des.oracle:meta-learning and feature selection for dynamic ensemble selection[J].Information Fusion,2017,38:84-103.
[4] Filchenkov A,Pendryak A.Datasets meta-feature description for recommending feature selection algorithm[C]∥Artificial Intelligence and Natural Language and Information Extraction,Social Media and Web Search Fruct Conference,IEEE,2015:11-18.
[5] Sousa A F M,Prudêncio R B C,Ludermir T B,et al.Active learning and data manipulation techniques for generating training examples in meta-learning[J].Neurocomputing,2016,194:45-55.
[6] Morais R F A B D,Miranda P B C,Silva R M A.A meta-learning method to select under-sampling algorithms for imbalanced data sets[C]∥Brazilian Conference on Intelligent Systems,IEEE Computer Society,2016:385-390.
[7] 曾子林,张宏军,张睿,等.基于元学习思想的算法选择问题综述[J].控制与决策,2014,29(6):961-968.
[8] Rossi A L D,Carvalho A C P L F,Soares C,et al.Meta stream:a meta-learning based method for periodic algorithm selection in time-changing data[J].Neurocomputing,2014,127(3):52-64.
[9] Song Qinbao,Wang Guangtao,Wang Chao.Automatic recommendation of classification algorithms based on data set characteristics[J].Pattern Recognition,2012,45(7):2672-2689.
[10] Pfahringer B,Bensusan H,Garrier C G.Meta-learning by landmarking various learning algorithms[C]∥Proc of the 17th Int Conf on Machine Learning,San Francisco:Morgan Kaufmann,2000:743-750.
[11] Peng Y H,Flach P A,Soares C,et al.Improved dataset characterization for meta-learning[C]∥Proc of Discovery Science 5th Int Conf Lubeck Germany,2002:141-152.
[12] Hilario M,Kalousis A.Fusion of meta-knowledge and meta-data for case-based model selection[C]∥Lecture Notes in Computer Science,2001,2168:180-191.
[13] Kalousis A,Hilario M.Building algorithm profiles for prior model selection in knowledge discovery systems[J].International Journal of Engineering Intelligent Systems for Electrical Engineering and Communications,2000,8(2):77-88.
[14] Brazdil P,Carrier C G,Soares C,et al.Meta learning:applications to data mining[M].Berlin:Springer Science and Business Media,2008.
[15] Spearman C.The proof and measurement of association between two things[J].The American Journal of Psychology,1904,15(1):72-101.
[16] Kendall M.Rank correlation methods[M].London:Charles Griffin and Company Limited,1948.

备注/Memo

备注/Memo:
收稿日期:2018-04-26
基金项目:国家自然科学基金(11501281),装备军内科研课题(面向作战任务的分队战斗体能数据分析评估系统建设)和江西省社科“十二五”规划课题(15GL44)资助项目.
通信作者:曾子林(1981-),女,江西鄱阳人,讲师,博士,主要从事元学习、特征选择方面的研究.E-Mail:zzljxnu@163.com
更新日期/Last Update: 2018-12-20