计算机科学 ›› 2017, Vol. 44 ›› Issue (Z11): 129-132.doi: 10.11896/j.issn.1002-137X.2017.11A.026
王蓉,刘遵仁,纪俊
WANG Rong, LIU Zun-ren and JI Jun
摘要: 传统的ID3决策树算法存在属性选择困难、分类效率不高、抗噪性能不强、难以适应大规模数据集等问题。针对该情况,提出一种基于属性重要度及变精度粗糙集的决策树算法,在去除噪声数据的同时保证了决策树的规模不会太庞大。利用多个UCI标准数据集对该算法进行了验证,实验结果表明该算法在所得决策树的规模和分类精度上均优于ID3算法。
[1] 梁凤兰.优化决策树改进挖掘算法仿真[J].计算机仿真,2013,30(11):264-267. [2] 张棪,曹健.面向大数据分析的决策树算法[J].计算机科学,2016,43(6A):374-379. [3] QUINLAN J R.Induction of Decision Trees[J].Machine Ler-ning,1986,1(1):81-106. [4] QUINLAN J R.Simplifying Decision Trees[J].InternationalJournal of Man-machine Studies,1987,7(3):221-234. [5] 洪家荣,丁明锋,李星原.一种新的决策树归纳学习算法[J].计算机学报,1995,18(6):470-474. [6] 刘小虎,李生.决策树的优化算法[J].软件学报,1998,9(10):797-800. [7] WANG S Q,WEI J M,YOU J P,et al.A VPRSM based approach for inducing decision trees[C]∥RSKT2006.Chongqing,China,2006:421-429. [8] 洪雪飞,徐维祥.基于变精度粗糙集的决策树改进方法[J].计算机工程与应用,2009,45(13):163-165. [9] 丁春荣,李龙澍.变精度粗糙集模型在决策树构造中的应用[J].计算机工程与科学,2010,32(7):86-88. [10] 鄂旭,任骏原,毕嘉娜,等.基于粗糙变精度的食品安全决策树研究[J].计算机技术与发展,2014,24(1):242-245. [11] BARANAUSKAS J A.The number of classes as a source for instability of decision tree algorithms in high dimensional datasets[J].Springer,2015,43(2):301-310. [12] LIANG C Q,ZHANG Y,SHI P,et al.Learning accurate very fast decision trees from uncertain data streams[J].Taylor & Francis,2015,46(16):3032-3050. [13] 王婧,王兴伟,赵悦.基于变精度粗糙集决策树垃圾邮件过滤[J].系统仿真学报,2016,28(3):705-710. [14] APNIK V.The nature of statistical learning theroy[M].New York:Springer,1995. [15] PAWLAK Z,SO-WINSKI R.Rough set approach to multi-attribute decision analysis[J].European Journal of Operational Research,1994,72(3):443-459. [16] LIU Y,HUANG W,JIANG Y,et al.Quick attribute reduct algorithm for neighborhood rough set model[J].Information Sciences,2014,271(7):65-81. [17] 娄畅,刘遵仁,郭功振.基于块集的邻域粗糙集的快速约简算法[J].计算机科学,2014,41(S2):337-339. |
No related articles found! |
|