参数独立的加权局部均值伪近邻分类算法

doi:10.11772/j.issn.1001-9081.2020091370

计算机应用 ›› 2021, Vol. 41 ›› Issue (6): 1694-1700.DOI: 10.11772/j.issn.1001-9081.2020091370

所属专题：数据科学与技术

参数独立的加权局部均值伪近邻分类算法

蔡瑞光, 张德生, 肖燕婷

西安理工大学理学院, 西安 710054

收稿日期:2020-09-07 修回日期:2020-10-24 出版日期:2021-06-10 发布日期:2020-11-10
通讯作者: 蔡瑞光
作者简介:蔡瑞光(1996-),女,陕西宝鸡人,硕士研究生,主要研究方向:数据挖掘、分类分析;张德生(1964-),男,陕西永寿人,教授,博士,主要研究方向:概率论与数理统计;肖燕婷(1981-),女,陕西户县人,副教授,博士,主要研究方向:概率论与数理统计。
基金资助:
国家自然科学基金青年科学基金资助项目（11801438）。

Parameter independent weighted local mean-based pseudo nearest neighbor classification algorithm

CAI Ruiguang, ZHANG Desheng, XIAO Yanting

Faculty of Science, Xi'an University of Technology, Xi'an Shaanxi 710054, China

Received:2020-09-07 Revised:2020-10-24 Online:2021-06-10 Published:2020-11-10
Supported by:
This work is partially supported by the Youth Program of National Natural Foundation of China (11801438).

摘要/Abstract

摘要： 针对局部均值伪近邻（LMPNN）算法对k值敏感且忽略了每个属性对分类结果的不同影响等问题，提出了一种参数独立的加权局部均值伪近邻分类（PIW-LMPNN）算法。首先，利用差分进化算法的最新变体——基于成功历史记录的自适应参数差分进化（SHADE）算法对训练集样本进行优化，从而得到最佳k值和一组与类别相关的最佳权重；其次，计算样本间的距离时赋予每类的每个属性不同的权重，并对测试集样本进行分类。在15个实际数据集上进行了仿真实验，并把所提算法与其他8种分类算法进行了比较，实验结果表明，所提算法的分类准确率和F1值分别最大提高了约28个百分点和23.1个百分点；同时Wilcoxon符号秩检验、Friedman秩方差检验以及Hollander-Wolfe两处理的比较结果表明，所提出的改进算法在分类精度以及k值选择方面相较其他8种分类算法具有明显优势。

关键词: 局部均值伪近邻算法, 特征权重, 优化模型, 基于成功历史记录的自适应参数差分进化, 参数自适应

Abstract: Aiming at the problem that the Local Mean-based Pseudo Nearest Neighbor (LMPNN) algorithm is sensitive to the value of k and ignores the different influence of different attributes on the classification results, a Parameter Independent Weighted Local Mean-based Pseudo Nearest Neighbor classification (PIW-LMPNN) algorithm was proposed. Firstly, the Success-History based parameter Adaptation for Differential Evolution (SHADE) algorithm, the latest variant of differential evolution algorithm, was used to optimize the training set samples to obtain the best k value and a set of best weights related to the classes. Secondly, when calculating the distance between samples, different weights were assigned to different attributes of different classes, and the test set samples were classified. Finally, simulations were performed on 15 real datasets and the proposed algorithm was compared to other eight classification algorithms. The results show that the proposed algorithm has the classification accuracy and F1 value increased by about 28 percentage points and 23.1 percentage points respectively. At the same time, the comparision results of Wilcoxon signed-rank test, Friedman rank variance test and Hollander-Wolfe's pairwise processing show that the proposed improved algorithm outperforms the other eight classification algorithms in terms of classification accuracy and k value selection.

Key words: Local Mean-based Pseudo Nearest Neighbor (LMPNN) algorithm, feature weighting, optimization model, Success-History based parameter Adaptation for Differential Evolution (SHADE), parameter adaption

中图分类号:

TP301

蔡瑞光, 张德生, 肖燕婷. 参数独立的加权局部均值伪近邻分类算法[J]. 计算机应用, 2021, 41(6): 1694-1700.

CAI Ruiguang, ZHANG Desheng, XIAO Yanting. Parameter independent weighted local mean-based pseudo nearest neighbor classification algorithm[J]. Journal of Computer Applications, 2021, 41(6): 1694-1700.

参考文献

[1] DUDA R O,HART P E,STORK D G. Pattern Classification[M]. New York:Wiley & Sons,Inc.,2000:6-10.
[2] MITANI Y,HAMAMOTO Y. A local mean-based nonparametric classifier[J]. Pattern Recognition Letters,2006,27(10):1151-1159.
[3] ZENG Y,YANG Y,ZHAO L. Pseudo nearest neighbor rule for pattern classification[J]. Expert Systems with Applications,2009, 36(2 Pt 2):3587-3595.
[4] GOU J, ZHAN Y, RAO Y, et al. Improved pseudo nearest neighbor classification[J]. Knowledge-Based Systems,2014,70:361-375.
[5] 葛月月, 曾勇, 胡江平, 等. 改进局部均值与类均值权重的近邻分类[J]. 计算机工程与应用, 2017, 53(17):137-142.(GE Y Y, ZENG Y,HU J P,et al. Improved nearest neighbor classification based on local mean and class mean with modified weights[J]. Computer Engineering and Applications, 2017, 53(17):137-142.)
[6] GOU J,QIU W,YI Z,et al. A local mean representation-based knearest neighbor classifier[J]. ACM Transactions on Intelligent Systems and Technology,2019,10(3):Article No. 29.
[7] GOU J,QIU W,YI Z,et al. Locality constrained representationbased k-nearest neighbor classification[J]. Knowledge-Based Systems,2019,167:38-52.
[8] 邱文默. 基于局部多均值的近邻模式分类[D]. 镇江:江苏大学, 2019:32-52.(QIU W M. Multi local means-based nearest neighbor pattern classification[D]. Zhenjiang:Jiangsu University,2019:32-52.)
[9] MA H,GOU J,WANG X,et al. Sparse coefficient-based k-nearest neighbor classification[J]. IEEE Access,2017,5:16618-16634.
[10] STORN R,PRICE K. Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of Global Optimization,1997,11(4):341-359.
[11] TANABE R,FUKUNAGA A. Success-history based parameter adaptation for differential evolution[C]//Proceedings of the 2013 IEEE Congress on Evolutionary Computation. Piscataway:IEEE, 2013:71-78.
[12] BISWAS N, CHAKRABORTY S, MULLICK S S, et al. A parameter independent fuzzy weighted k-nearest neighbor classifier[J]. Pattern Recognition Letters,2018,101:80-87.
[13] NABABAN A A,SITOMPUL O S,TULUS. Attribute weighting based K-nearest neighbor using gain ratio[J]. Journal of Physics:Conference Series,2018,1007:Article No. 012007.
[14] 茆诗松, 吕晓玲. 概率论与数理统计教程[M]. 2版. 北京:中国人民大学出版社,2016:415-420. (MAO S S,LYU X L. Mathematical Statics[M]. 2nd ed. Beijing:China Renmin University Press,2016:415-420.)
[15] FRIEDMAN M. The use of ranks to avoid the assumption of normality implicit in the analysis of variance[J]. Journal of the American Statistical Association,1937,32(200):675-701.
[16] 王星, 褚挺进. 非参数统计[M]. 2版. 北京:清华大学出版社, 2014:126-131.(WANG X,CHU T J. Non-parametric Statistics[M]. 2nd ed. Beijing:Tsinghua University Press, 2014:126-131.)
[17] GOU J,QIU W,MAO Q,et al. A multi-local means based nearest neighbor classifier[C]//Proceedings of the 2017 IEEE 29th International Conference on Tools with Artificial Intelligence. Piscataway:IEEE,2017:448-452.

参数独立的加权局部均值伪近邻分类算法

Parameter independent weighted local mean-based pseudo nearest neighbor classification algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

[1]	杨华龙, 王美玉, 辛禹辰. 横向整合战略下异质车辆库存路径优化模型[J]. 计算机应用, 2021, 41(10): 3040-3048.
[2]	方宏道, 周颖玥, 林茂松. 基于贝叶斯非局部平均滤波的超声图像斑点噪声抑制算法[J]. 计算机应用, 2018, 38(3): 848-853.
[3]	王超, 王坚. 基于地面场粒子群优化算法的高密度人群应急疏散建模[J]. 计算机应用, 2017, 37(12): 3597-3601.
[4]	程铃钫, 杨天鹏, 陈黎飞. 不平衡数据的软子空间聚类算法[J]. 计算机应用, 2017, 37(10): 2952-2957.
[5]	束珏, 成卫青, 邓聪. 基于话题标签和转发关系的微博聚类和主题词提取[J]. 计算机应用, 2016, 36(2): 460-464.
[6]	龙鹏, 鲁华祥. LoG边缘算子改进的加权引导滤波算法[J]. 计算机应用, 2015, 35(9): 2661-2665.
[7]	闫华, 高黎, 刘国勇, 王红旗. 基于多时间窗的油料保障模型[J]. 计算机应用, 2015, 35(7): 2096-2100.
[8]	田强夏永滢付晓东李昌志王威. 基于成本效益系数的Web服务组合可靠性优化方法[J]. 计算机应用, 2014, 34(3): 683-689.
[9]	马雯雯邓一贵. 新的短文本特征权重计算方法[J]. 计算机应用, 2013, 33(08): 2280-2282.
[10]	代志华付晓东黄袁贾楠. 基于最大熵原理的Web服务QoS概率分布获取[J]. 计算机应用, 2012, 32(10): 2728-2731.
[11]	巴黎杨明顺高新勤武晓青. 考虑运输因素的网络化制造资源优化配置方法[J]. 计算机应用, 2010, 30(11): 2902-2905.
[12]	罗永红陈特放张友生. 基于数据网格的分布式查询优化模型[J]. 计算机应用, 2008, 28(10): 2553-2557.
[13]	孙靖林杰 . 基于蚁群算法的MC供应链调度优化研究[J]. 计算机应用, 2006, 26(11): 2631-2634.