摘要
在直推式支持向量机中,迭代过程中样本标注错误会导致错误传递,影响下一次迭代中样本标注准确度,使得错误不断的被积累,造成最终分类超平面的偏移,另外在传统单个分类器下,提高样本标注准确度与提高算法训练速度之间是矛盾的,无法得到兼顾.针对此,本文把投票机制和协同思想引入到直推式支持向量机中,提出一种协同标注的直推式支持向量机算法,利用多个分类器的投票结果对样本进行标注,提高样本标注的准确度,利用多个分类器进行协同训练提高算法的训练速度.最后实验结果表明,所提出算法能够利用投票机制和协同思想提高最终分类器的分类精度和算法的训练速度.
In transductive support vector machine, sample labeling error will result in error propagation in the iterative process. It affects the accuracy of sample labeling in the next iteration and makes mistakes constantly being accumulated. Eventually leading to classifica- tion hyperplane offset. In addition,under the traditional single classifier,to improve the accuracy of the sample labeling and to improve the speed of the algorithm is contradictory,cannot take into account. In this paper, we introduce the voting mechanism and the coopera- tive idea into the transductive support vector machine, and propose an algorithm of transductive support vector machine based on coop- erative labeling. The voting results of multiple classifiers are used to label the samples, which improve the accuracy of sample labeling. Use multiple classifiers for cooperative training to improve the training speed of the algorithm. Finally, the experimental results show that the proposed algorithm can improve the classification accuracy and the training speed by using the voting mechanism and coopera- tive idea.
出处
《小型微型计算机系统》
CSCD
北大核心
2016年第11期2443-2447,共5页
Journal of Chinese Computer Systems
基金
陕西省自然科学基础研究计划项目(2015JM6347)资助
陕西省教育厅专项科研计划项目(15JK1218)资助
商洛学院科学与技术研究项目(15sky010)资助
关键词
支持向量机
直推式学习
半监督学习
协同标注
support vector Machine
transductive learning
semi-supervised learning
cooperative labeling
