代价敏感相关向量机

计算机与现代化

代价敏感相关向量机

南京航空航天大学计算机科学与技术学院，江苏南京210016

收稿日期:2014-11-12 出版日期:2015-02-28 发布日期:2015-03-06
作者简介:作者简介：苏乐群（1990），男，安徽宿松人，南京航空航天大学计算机科学与技术学院硕士研究生，研究方向：模式识别；冯爱民（1971），女，江苏南京人，副教授，硕士生导师，博士，研究方向：模式识别，机器学习，系统结构。
基金资助:
国家自然科学基金资助项目（61170152）

Cost Sensitive Relevance Vector Machine

College of Computer Science & Technology， Nanjing University Aeronautics ＆ Astronautics， Nanjing 210016， China

Received:2014-11-12 Online:2015-02-28 Published:2015-03-06

摘要/Abstract

摘要：

相关向量机（RVM）是在稀疏贝叶斯框架下提出的稀疏模型，由于其强大的稀疏性和泛化能力，近年来在机器学习领域得到了广泛研究和应用，但和传统的决策树、神经网络算法及支持向量机一样

，RVM不具有代价敏感性，不能直接用于代价敏感学习。针对监督学习中错误分类带来的代价问题，提出代价敏感相关向量分类（CSRVC）算法，在相关向量机的基础上，通过赋予每类样本不同的误分

代价，使其更加注重误分类代价较高的样本分类准确率，使得整体误分类代价降低以实现代价敏感挖掘。实验结果表明，该算法具有良好的稀疏性并能够有效地解决代价敏感分类问题。

关键词: 相关向量机（RVM）, 代价敏感, 代价敏感相关向量分类

Abstract:

Relevance Vector Machine (RVM) is a sparse model proposed on the basis of sparse Bayesian framework, it has been widely studied and applied in the field of machine

learning in recent years because of its strong sparsity and generalization ability. However, like the traditional decision tree, neural network algorithm and support vector

machine, RVM does not have the expense of sensitivity, can not be directly used for costsensitive learning. To deal with the cost sensitive problem brought by

misclassification in supervised learning, costsensitive relevance vector classification(CSRVC) algorithm was proposed by integrating misclassification cost of each type

sample based on RVM. Experiments show that CSRVC has good sparsity and can effectively solve the problem of costsensitive classification.


Key words: relevance vector machine(RVM), cost sensitive, CSRVC

中图分类号:

TP301.6

苏乐群，冯爱民. 代价敏感相关向量机[J]. 计算机与现代化.

SU Lequn， FENG Aimin. Cost Sensitive Relevance Vector Machine[J]. Computer and Modernization.

［1］
叶志飞,文益民,吕宝粮. 不平衡分类问题研究综述［J］. 智能系统学报, 2009,4(2):148156.
［2］Habbema J D F, Hermans J, Van Der Burgt A T. Cases of doubt in allocation problems［J］. Biometrika, 1974,61(2):313324.
［3］Chawla N V, Bowyer K W, Hall L O, et al. SMOTE:Synthetic minority oversampling technique［J］. Journal of Artificial Intelligence Research, 2002,16:321357.
［4］Domingos P Metacost. A general method for making classifiers costsensitive［C］// Proceedings of the fifth ACM SIGKDD International Conference on Knowledge Discovery and

Data Mining. ACM, 1999:155164.
［5］Breiman L. Bagging predictors［J］. MachineLearning, 1996,24(2):123140.
［6］Brefeld U, Geibel P, Wysotzki F. Support vector machines with example dependent costs［M］// Machine Learning: ECML 2003. Springer Berlin Heidelberg, 2003:2334.
［7］刘胥影,姜远,周志华. 类别不平衡性对代价敏感学习的影响［C］// 人工智能学会第12届学术年会,中国人工智能进展. 2007:4550.
［8］郑恩辉,李平,宋执环. 代价敏感支持向量机［J］. 控制与决策, 2006,21(4):473476.
［9］Fumera G, Roli F. Costsensitive learning in support vector machines［C］// Workshop on Machine Learning, Methods and Applications, Held in the Context of the 8th Meeting

of the Italian Association of Artificial Intelligence. 2002.
［10］曹莹,苗启广,刘家辰,等. 具有Fisher一致性的代价敏感Boosting算法［J］. 软件学报, 2013,24(11):25842596.
［11］阮晓宏,黄小猛,袁鼎荣,等. 基于异构代价敏感决策树的分类器算法［J］. 计算机科学, 2013,40(11A):140142.
［12］Bishop C M, Tipping M E. Variational relevance vector machines［C］// Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence. 2000:4653.
［13］Vapnik V N, Vapnik V. StatisticalLearning Theory［M］. New York: Wiley, 1998.
［14］Li Y, Campbell C, Tipping M. Bayesian automatic relevance determination algorithms for classifying gene expression data［J］. Bioinformatics, 2002,18(10):13321339.
［15］Mackay D. The evidence framework applied to classification networks［J］. Neural Computation, 1992,4(5):720736
［16］Thayananthan A. Relevance Vector Machine based Mixture of Experts［D］. Department of Engineering, University of Cambridge, 2005.
［17］Platt J C. Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods［M］// Advances in Large Margin Classifiers. MIT Press,

1999:6174.
［18］Chen H, Tino P,Yao X. Probabilistic classification vector machines［J］. IEEE Transactions on Neural Networks, 2009,20(6):901914.
［19］Han J, Kamber M. Data Mining,Southeast Asia Edition: Concepts and Techniques［M］. Morgan Kaufmann, 2006.

［20］周志华. 普适机器学习［EB/OL］. http://wenku.baidu.com/link?url=ZhAh423i0RLQfcCUKaTPk0ojoV5765Zu7O0vWxCkLlgjF8YmJGzmNKkoiP1hjhD6zeOmXvmaag0we

MRwHbbPYHBEAlVooCRLBfeL3EVhDa, 20141112.

[1]	周传华, 朱俊杰, 徐文倩, 邓佳佳. 基于聚类欠采样的集成分类算法[J]. 计算机与现代化, 2021, 0(11): 72-76.
[2]	席晓燕,于化龙. #br# 耦合样本先验分布信息的加权极限学习机[J]. 计算机与现代化, 2018, 0(08): 61-.
[3]	许智彪. 基于代价敏感主动学习算法的2型糖尿病诊断[J]. 计算机与现代化, 2018, 0(06): 84-.