基于RNN集成学习的个人轨迹恢复方法
|
摘要
从多个轨迹数据库中连接并恢复出较为完整的个人轨迹对出行推荐和移动导航具有重要的意义。基于个人轨迹恢复,提出RNN集成学习方法。定义个人轨迹恢复的形式化模型,利用轨迹点数目采样模式将每个训练库划分为多个训练子库,并采用RNN网络模型描述个人轨迹的可拼接程度,使用集成学习方法构建多个RNN网络,以达到恢复个人轨迹的目的。实验结果表明,该方法可以较好地捕获轨迹时空连续性特征,实现个人轨迹恢复。
Connecting and restoring a relatively complete personal trajectory from multiple trajectory databases is important for travel recommendations and mobile navigation.Based on personal trajectory recovery,an integrated learning method for RNN is proposed.By defining a formal model of personal trajectory recovery,each training library is divided into multiple training sub-libraries by using the track point number sampling mode,and an RNN network model is used to describe the splicing degree of the personal trajectory.An integrated learning method is used to construct multiple RNN networks to achieve the goal of restoring personal trajectories.Experimental results show that this method can capture the temporal and spatial continuity of the trajectory and realize the personal trajectory recovery.
Connecting and restoring a relatively complete personal trajectory from multiple trajectory databases is important for travel recommendations and mobile navigation.Based on personal trajectory recovery,an integrated learning method for RNN is proposed.By defining a formal model of personal trajectory recovery,each training library is divided into multiple training sub-libraries by using the track point number sampling mode,and an RNN network model is used to describe the splicing degree of the personal trajectory.An integrated learning method is used to construct multiple RNN networks to achieve the goal of restoring personal trajectories.Experimental results show that this method can capture the temporal and spatial continuity of the trajectory and realize the personal trajectory recovery.
引文
[1] SHANG S,GUO D,LIU J,et al.Human mobility prediction and unobstructed route planning in public transport networks[C]//Proceedings of the 15th International Conference on Mobile Data Management.Washington,D.C.,USA:IEEE Computer Society,2014:43-48.
[2] WU G,DING Y,LI Y,et al.Mining spatio-temporal reachable regions over massive trajectory data[C]//Proceedings of the 33rd International Conference on Data Engineering.Washington D.C.,USA:IEEE Press,2017:1283-1294.
[3] SIQUEIRA F D L,PLENTZ P D M,PIERI E R D.Semantic trajectory applied to the navigation of autonomous mobile robots[C]//Proceedings of the 13th International Conference of Computer Systems and Applications.Washington D.C.,USA:IEEE Press,2017.
[4] BHUNIA A K,BHOWMICK A,BHUNIA A K,et al.Handwriting trajectory recovery using end-to-end deep encoder-decoder network [EB/OL].[2017-12-05].https://arxiv.org/ftp/arxiv/papers/1801/1801.07211. pdf.
[5] NOUBIGH Z,KHERALLAH M.A survey on handwriting recognition based on the trajectory recovery technique[C]//Proceedings of International Workshop on Arabic Script Analysis and Recognition.Washington D.C.,USA:IEEE Press,2017:69-73.
[6] LI X,SHU W,LI M,et al.Performance evaluation of vehicle-based mobile sensor networks for traffic monitoring[J].IEEE Transactions on Vehicular Technology,2009,58(4):1647-1653.
[7] TONG C,CHEN H,QI X,et al.A framework for bus trajectory extraction and missing data recovery for data sampled from the internet[J].Sensors,2017,17(2):342.
[8] WU H,XUE M,CAO J,et al.Fuzzy trajectory linking[C]//Proceedings of the 32nd International Conference on Data Engineering.Washington D.C.,USA:IEEE Press,2016:859-870.
[9] 梁天新,杨小平,王良,等.记忆神经网络的研究与发展[J].软件学报,2017,28(11):2905-2924.
[10] 韩冬煦,常宝宝.中文分词模型的领域适应性方法[J].计算机学报,2015,38(2):272-281.
[11] KACHERGIS G,YU C,SHIFFRIN R M.A bootstrapping model of frequency and context effects in word learning[J].Cognitive Science,2016,41(3):590-622.
[12] ZHAO Y,YANG R,CHEVALIER G,et al.Applying deep bidirectional LSTM and mixture density network for basketball trajectory prediction[J].Optik,2018,158:266-272.
[13] ZHOU Z H.Ensemble methods:foundations and algorithms[M].[S.l.]:CRC Press,2012:68-70.
[14] 付忠良.通用集成学习算法的构造[J].计算机研究与发展,2013,50(4):861-872.
[15] 李凯,崔丽娟.集成学习算法的差异性及性能比较[J].计算机工程,2008,34(6):35-37.
[16] T-Drive trajectory data sample[EB/OL].[2017-12-05].http://research.microsoft.com/apps/pubs/?id=152883.
[2] WU G,DING Y,LI Y,et al.Mining spatio-temporal reachable regions over massive trajectory data[C]//Proceedings of the 33rd International Conference on Data Engineering.Washington D.C.,USA:IEEE Press,2017:1283-1294.
[3] SIQUEIRA F D L,PLENTZ P D M,PIERI E R D.Semantic trajectory applied to the navigation of autonomous mobile robots[C]//Proceedings of the 13th International Conference of Computer Systems and Applications.Washington D.C.,USA:IEEE Press,2017.
[4] BHUNIA A K,BHOWMICK A,BHUNIA A K,et al.Handwriting trajectory recovery using end-to-end deep encoder-decoder network [EB/OL].[2017-12-05].https://arxiv.org/ftp/arxiv/papers/1801/1801.07211. pdf.
[5] NOUBIGH Z,KHERALLAH M.A survey on handwriting recognition based on the trajectory recovery technique[C]//Proceedings of International Workshop on Arabic Script Analysis and Recognition.Washington D.C.,USA:IEEE Press,2017:69-73.
[6] LI X,SHU W,LI M,et al.Performance evaluation of vehicle-based mobile sensor networks for traffic monitoring[J].IEEE Transactions on Vehicular Technology,2009,58(4):1647-1653.
[7] TONG C,CHEN H,QI X,et al.A framework for bus trajectory extraction and missing data recovery for data sampled from the internet[J].Sensors,2017,17(2):342.
[8] WU H,XUE M,CAO J,et al.Fuzzy trajectory linking[C]//Proceedings of the 32nd International Conference on Data Engineering.Washington D.C.,USA:IEEE Press,2016:859-870.
[9] 梁天新,杨小平,王良,等.记忆神经网络的研究与发展[J].软件学报,2017,28(11):2905-2924.
[10] 韩冬煦,常宝宝.中文分词模型的领域适应性方法[J].计算机学报,2015,38(2):272-281.
[11] KACHERGIS G,YU C,SHIFFRIN R M.A bootstrapping model of frequency and context effects in word learning[J].Cognitive Science,2016,41(3):590-622.
[12] ZHAO Y,YANG R,CHEVALIER G,et al.Applying deep bidirectional LSTM and mixture density network for basketball trajectory prediction[J].Optik,2018,158:266-272.
[13] ZHOU Z H.Ensemble methods:foundations and algorithms[M].[S.l.]:CRC Press,2012:68-70.
[14] 付忠良.通用集成学习算法的构造[J].计算机研究与发展,2013,50(4):861-872.
[15] 李凯,崔丽娟.集成学习算法的差异性及性能比较[J].计算机工程,2008,34(6):35-37.
[16] T-Drive trajectory data sample[EB/OL].[2017-12-05].http://research.microsoft.com/apps/pubs/?id=152883.