摘要
为提高短时交通状态预测的精度,使交通管理者更有效地进行交通规划和管理,本文把基于L 1范数距离度量的最小二乘孪生有界支持向量机(twin bounded support vector machine,TBSVM)扩展成多分类算法用于短时交通状态预测,简称MLSTBSVM L1.在实验数据上对MLSTBSVM L1算法的有效性进行验证,实验结果表明,相比于其他预测算法,提出的MLSTBSVM L1算法在预测精度上有较大提升.
To improve the accuracy of short-term traffic condition prediction and make planning and management more effective for traffic managers,the least squares Twin Bounded Support Vector Machine(TBSVM)based on L 1-norm distance is extended to a new algorithm(MLSTBSVM L1)which could solve multi-classification problems.The validity of the proposed MLSTBSVM L1 is verified through experiments and the results demonstrate that the MLSTBSVM L1 algorithm has significant improvement in prediction accuracy compared with other prediction algorithms.
作者
闫贺
朱丽
戚湧
Yan He;Zhu Li;Qi Yong(School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing 210094,China;School of Transportation,Southeast University,Nanjing 210096,China;JSTI Group Co.,Ltd,Planning Institute,Nanjing 210019,China)
出处
《南京师大学报（自然科学版）》
CAS
CSCD
北大核心
2019年第3期129-137,共9页
Journal of Nanjing Normal University(Natural Science Edition)
基金
国家重点研究计划政府间国际科技创新合作重点专项(2016YFE0108000)
国家重点自然科学基金项目(51238008)
国家自然科学基金(61272419、61772273)
江苏省自然科学基金(BK20141403)
2018江苏省普通高校学术学位研究生科研创新计划项目(KYCX18_0424)
