CAJ | 학술논문

海量的交通流数据中一定隐藏着某些潜在的交通状态演变规律，然而少有研究能以实验的方法支持这种观点。本文提出一种基于频繁序列匹配的交通状态趋势预测方法：首先，介绍交通状态序列模型及序列切分处理，其次，提出投影压缩序列相似度计算方法及序列匹配算法，第三，讨论基于频繁序列匹配的交通序列预测算法。通过真实采集数据验证本文提出的方法可行有效。
해량적교통류수거중일정은장착모사잠재적교통상태연변규률，연이소유연구능이실험적방법지지저충관점。본문제출일충기우빈번서렬필배적교통상태추세예측방법：수선，개소교통상태서렬모형급서렬절분처리，기차，제출투영압축서렬상사도계산방법급서렬필배산법，제삼，토론기우빈번서렬필배적교통서렬예측산법。통과진실채집수거험증본문제출적방법가행유효。
As known that some latent laws of traffic status evolvement must exist among huge traffic data. However, little work has been done to support this view by experimental methods. In this paper, a new approach is presented to investigate the traffic flow prediction on the basis of traffic status frequent sequence similarity. Firstly, the concepts of traffic status frequent sequences and the algorithm of sequence similarity were introduced. Secondly, the measure of frequent sequence matching was proposed. Thirdly, the traffic status prediction was discussed on the basis of similarity matching. Lastly, the efficiency and validity of this method was proved by real experiment data observed.