计算机科学 ›› 2021, Vol. 48 ›› Issue (11A): 191-197.doi: 10.11896/jsjkx.201200015

• 大数据&数据科学 • 上一篇    下一篇

基于自适应时间戳与多尺度特征提取的轨迹下一足迹预测模型

李艾玲, 张凤荔, 高强, 王瑞锦   

  1. 电子科技大学信息与软件工程学院网络与数据安全四川省重点实验室 成都610054
  • 出版日期:2021-11-10 发布日期:2021-11-12
  • 通讯作者: 张凤荔(fzhang@uestc.edu.cn)
  • 作者简介:li.ailing.1225@gmail.com
  • 基金资助:
    国家自然科学基金(61802033,61472064,61602096);四川省科技计划(2018GZ0087,2019YJ0543);博士后基金项目(2018M643453);广东省国家重点实验室项目(2017B030314131);网络与数据安全四川省重点实验室开放课题(NDSMS201606)

Trajectory Next Footprint Prediction Model Based on Adaptive Timestamp and Multi-scale Feature Extraction

LI Ai-ling, ZHANG Feng-li, GAO Qiang, WANG Rui-jin   

  1. Network and Data Security Key Laboratory of Sichuan Province,School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 610054,China
  • Online:2021-11-10 Published:2021-11-12
  • About author:LI Ai-ling,born in 1995,postgraduate.Her main research interests include machine learning,data mining,pattern mining and software development techniques.
    ZHANG Feng-li,born in 1963,Ph.D,professor,is a member of China Computer Federation.Her main research interests include network security and network engineering,cloud computing and big data and machine learning.
  • Supported by:
    National Natural Science Foundation of China(61802033,61472064,61602096),Science and Technology Project of Sichuan Province,China(2018GZ0087,2019YJ0543),Postdoctoral Fund Project(2018M643453),Guangdong State Key Laboratory Project(2017B030314131) and Sichuan Key Laboratory of Network and Data Security Open Class(NDSMS201606).

摘要: 基于位置的服务已经成为人类生活方式的一部分,各种移动终端设备产生了大量时空上下文用户信息,其可被用于预测用户的下一个足迹。目前已提出一些解决方案来预测用户下一个足迹,包括递归运动函数(RMF)、矩阵分解(MF)、差分自回归移动平均模型(ARIMA)、马尔可夫链(MC)、个性化马尔可夫链(FPMC)、卡尔曼滤波器(KF)、高斯混合模型和张量分解(TF)。除此之外,也可以使用诸如ST-RNN,POI2Vec,DeepMove,VANext等深度神经网络方法来预测用户的下一个足迹,这些方法利用递归神经网络(RNN)捕获来自人类活动的顺序运动模式。然而,现有方法使用一些人为设定的阈值来分割人类移动性数据以进行用户运动模式学习,人为固定时间戳设置不仅引入了人为主观因素,而且忽略了不同用户之间的差异性,这可能会导致移动模式发生偏差;而且现有方法针对用户轨迹特征提取过于单一化,单一特征忽略了很多用户轨迹潜在信息。基于自适应时间戳与多尺度特征提取的轨迹预测模型(AMSNext)旨在首次结合历史轨迹数据的时间统计特性,自适应地为每一个用户定义个性化时间戳,关注不同用户运动模式之间的差异性;并结合时间序列特征提取方法多尺度对用户轨迹特征进行提取,同时为实现多尺度特征量纲统一,将会采取归一化因果嵌入对特征进行向量嵌入。实验证明,该模型可以取得较高的预测精度。

关键词: 归一化嵌入, 轨迹预测, 时间序列, 特征提取, 自适应时间戳

Abstract: Location-based services have become a part of human life style,and various mobile terminal devices generate a large amount of temporal and spatial contextual user information,which can be used to predict the user's next footprint.Some solutions have been proposed to predict the user's next footprint,including recursive motion function (RMF),matrix factorization (MF),differential autoregressive moving average model (ARIMA),Markov chain (MC),and personalization Markov chain (FPMC),Kalman filter (KF),Gaussian mixture model and tensor decomposition (TF).In addition,deep neural network methods such as ST-RNN,POI2Vec,DeepMove,VANext,etc.can also be used to predict the user's next footprint.These methods use recurrent neural networks (RNN) to capture sequential motion patterns from human activities.However,existing methods use some artificially set thresholds to segment human mobility data for user movement pattern learning.The artificial fixed time stamp setting not only introduces human subjective factors,but also ignores the differences between different users.It may lead to deviations in the movement pattern.The existing methods for the extraction of user trajectory features are too singular,and a single feature ignores a lot of potential user trajectory information.The trajectory prediction model based on adaptive timestamp and multi-scale feature extraction (AMSNext) aims to combine the time statistical characteristics of historical trajectory data for the first time,adaptively define a personalized timestamp for each user,and focus on the differences between different user motion modes.Combined with time series feature extraction methods to extract user trajectory features at multiple scales,and at the same time,to achieve multi-scale feature dimension unification,normalized causal embedding will be used to embed features in vector.Experiments show that the model can achieve higher prediction accuracy.

Key words: Adaptive timestamp, Feature extraction, Normalized embedding, Time series, Trajectory prediction

中图分类号: 

  • TP181
[1]LIAN D F,ZHAO C,XIE X,et al.GeoMF:Joint Geographical Modeling and Matrix Factorization for Point-of-Interest Recommendation[C]//Acm Sigkdd International Conference on Knowledge Discovery & Data Mining.ACM,2014.
[2]GAMBS S,KILLIJIAN M O,DEL PRADO CORTEZ M N.Next Place Prediction Using Mobility Markov Chains[C]//MPM.2012.
[3]ASAHARA A,SATO A,MARUYAMA K,et al.Pedestrian-movement prediction based on mixed Markov-chain model[C]//Proceedings of the 19th International Conference on Advances in Geographic Information Systems.IL,USA,2011:25-33.
[4]ASHBROOK D,STARNER T.Learning significant locationsand predicting user movement with GPS[C]//Proceedings of the 6th International Symposium on Wearable Computers.Sardina,Italy,2003:275-286.
[5]VITTER J S,KRISHNAN P.Optimal prefetching via data compression[J].Journal of the ACM,NY,USA,1996,43:771-793.
[6]ZHENG W C,CAO B,ZHENG Y,et al.Collaborative Filtering Meets Mobile Recommendation:A User-centered Approach[C]//AAAI.2010.
[7]FENG S S,LI X T,ZENG Y F,et al.Personalized ranking metric embedding for next new POI recommendation[C]//International Conference on Artificial Intelligence.AAAI Press,2015.
[8]CHENG C,YANG H Q,LYU M R,et al.Where You Like to Go Next:Successive Point-of-Interest Recommendation[C]//IJCAI.2013.
[9]LI X T,CONG G,LI X L,et al.Rank-GeoFM:A Ranking BasedGeographical Factorization Method for Point of Interest Recommendation[C]//SIGIR.2015.
[10]RENDLE S,FREUDENTHALER C,SCHMIDT-THIEME L.Factorizing personalized Markov chains for next-basket recommendation[C]//Proceedings of the 19th International Conference on World Wide Web(WWW 2010).Raleigh,North Carolina,USA:ACM,2010:26-30.
[11]MIKOLOV T,CHEN K,CORRADO G,et al.Efficient Estimation of Word Representations in Vector Space[C]//ICLR 2013.2013.
[12]FENG S S,CONG G,AN B,et al.POI2Vec:Geographical Latent Representation for Predicting Future Visitors[C]//AAAI.2017.
[13]LIU Q,WU S,WANG L,et al.Predicting the Next Location:A Recurrent Model with Spatial and Temporal Contexts[C]//Thirtieth Aaai Conference on Artificial Intelligence.AAAI Press,2016.
[14]FENG J,LI Y,ZHANG C,et al.DeepMove:Predicting Human Mobility with Attentional Recurrent Networks[C]//WWW.2018.
[15]GAO Q,ZHOU F,TRAJCEVSKI G,et al.Predicting HumanMobility via Variational Attention[C]//WWW.2019.
[16]ZHOU F,GAO Q,ZHANG K P,et al.Trajectory-User Linking via Variational AutoEncoder[C]//IJCAI.2018.
[17]YUAN J,ZHENG Y,ZHANG L,et al.Where to Find My Next Passenger?[C]//UbiComp 2011:Ubiquitous Computing,13th International Conference(UbiComp 2011).Beijing,China,2011:17-21.
[18]YUAN,JING N,ZHENG,et al.1 T-Finder:A RecommenderSystem for Finding Passengers and Vacant Taxis[J].IEEE Transactions on Knowledge and Data Engineering,2012,25(10):2390-2403.
[19]ZHENG Y,ZHANG L Z,XIE X,et al.Mining interesting locations and travel sequences from GPS trajectories[C]//WWW.2009.
[20]BAO J,ZHENG Y,MOKBEL M F.Location-based and prefe-rence-aware recommendation using sparse geosocial networkingdata[C]//ACM SIGSPATIAL GIS.2012.
[21]ZHENG Y,ZHANG L Z,XIE X,et al.Mining correlation between locations using human location history[C]//ACM SIGSPATIAL GIS.2009.
[22]ZHENG Y,XIE X.Learning location correlation from GPS trajectories[C]//MDM.2010.
[23]ZHENG Y,XIE X.Learning travel recommendations from user-generated GPS traces[J].Acm Transactions on Intelligent Systems & Technology,2011,2(1):1-29.
[24]GAO Q,ZHANG F,YAO F,et al.Adversarial Mobility Learning for Human Trajectory Classification[J].IEEE Access,2020,PP(99):1-1.
[25]YANG D Q,ZHANG D Q,ZHENG V W,et al.Modeling user activity preference by leveraging user spatial temporal characteristics in LBSNs[J].IEEE Transactions on Systems,Man,and Cybernetics:Systems,2015,45(1):129-142.
[26]GAO Q,ZHANG F L,WANG R J,et al.Trajectory Big Data:A Review of Key Technologies in Data Processing Summary[J].Ruan Jian Xue Bao/Journal of Software,2017,28(4):959-992.
[27]XU J J,ZHENG K,CHI M M,et al.Trajectory big data:data,application and technology status[J].Journal on Communications,2015,36(12):97-105.
[28]MAO J L,JIN C Q,ZHANG Z G,et al.Anomaly Detection for Trajectory big data:Advancements and Framework[J].Ruan Jian Xue Bao,2017,28(1):17-34.
[29]MENG X W,LI R C,ZHANG Y J,et al.Survery on mobile rec-ommender systems based on user trajectory data[J].Ruan Jian Xue Bao,2018,29(10):3111-3133.
[30]PENG H W,JIN Y Y,LU X Q,et al.Context-aware POI Recommendation Based on Matrix Factorization [J].Chinese Journal of Computers,2019(8):1797-1811.
[31]ZHANG J L,SHI H L,CUI L.Location Prediction Model Based on Transportation Mode and Semantic Trajectory[J].Computer Research and Development,2019,56(7):1357-1369.
[1] 张源, 康乐, 宫朝辉, 张志鸿.
基于Bi-LSTM的期货市场关联交易行为检测方法
Related Transaction Behavior Detection in Futures Market Based on Bi-LSTM
计算机科学, 2022, 49(7): 31-39. https://doi.org/10.11896/jsjkx.210400304
[2] 高振卓, 王志海, 刘海洋.
嵌入典型时间序列特征的随机Shapelet森林算法
Random Shapelet Forest Algorithm Embedded with Canonical Time Series Features
计算机科学, 2022, 49(7): 40-49. https://doi.org/10.11896/jsjkx.210700226
[3] 曾志贤, 曹建军, 翁年凤, 蒋国权, 徐滨.
基于注意力机制的细粒度语义关联视频-文本跨模态实体分辨
Fine-grained Semantic Association Video-Text Cross-modal Entity Resolution Based on Attention Mechanism
计算机科学, 2022, 49(7): 106-112. https://doi.org/10.11896/jsjkx.210500224
[4] 程成, 降爱莲.
基于多路径特征提取的实时语义分割方法
Real-time Semantic Segmentation Method Based on Multi-path Feature Extraction
计算机科学, 2022, 49(7): 120-126. https://doi.org/10.11896/jsjkx.210500157
[5] 刘伟业, 鲁慧民, 李玉鹏, 马宁.
指静脉识别技术研究综述
Survey on Finger Vein Recognition Research
计算机科学, 2022, 49(6A): 1-11. https://doi.org/10.11896/jsjkx.210400056
[6] 刘宝宝, 杨菁菁, 陶露, 王贺应.
基于DE-LSTM模型的教育统计数据预测研究
Study on Prediction of Educational Statistical Data Based on DE-LSTM Model
计算机科学, 2022, 49(6A): 261-266. https://doi.org/10.11896/jsjkx.220300120
[7] 高元浩, 罗晓清, 张战成.
基于特征分离的红外与可见光图像融合算法
Infrared and Visible Image Fusion Based on Feature Separation
计算机科学, 2022, 49(5): 58-63. https://doi.org/10.11896/jsjkx.210200148
[8] 左杰格, 柳晓鸣, 蔡兵.
基于图像分块与特征融合的户外图像天气识别
Outdoor Image Weather Recognition Based on Image Blocks and Feature Fusion
计算机科学, 2022, 49(3): 197-203. https://doi.org/10.11896/jsjkx.201200263
[9] 高堰泸, 徐圆, 朱群雄.
基于A-DLSTM夹层网络结构的电能消耗预测方法
Predicting Electric Energy Consumption Using Sandwich Structure of Attention in Double -LSTM
计算机科学, 2022, 49(3): 269-275. https://doi.org/10.11896/jsjkx.210100006
[10] 任首朋, 李劲, 王静茹, 岳昆.
基于集成回归决策树的lncRNA-疾病关联预测方法
Ensemble Regression Decision Trees-based lncRNA-disease Association Prediction
计算机科学, 2022, 49(2): 265-271. https://doi.org/10.11896/jsjkx.201100132
[11] 张师鹏, 李永忠.
基于降噪自编码器和三支决策的入侵检测方法
Intrusion Detection Method Based on Denoising Autoencoder and Three-way Decisions
计算机科学, 2021, 48(9): 345-351. https://doi.org/10.11896/jsjkx.200500059
[12] 冯霞, 胡志毅, 刘才华.
跨模态检索研究进展综述
Survey of Research Progress on Cross-modal Retrieval
计算机科学, 2021, 48(8): 13-23. https://doi.org/10.11896/jsjkx.200800165
[13] 张丽倩, 李孟航, 高珊珊, 张彩明.
面向计算机辅助舌诊关键问题的解决方案综述
Summary of Computer-assisted Tongue Diagnosis Solutions for Key Problems
计算机科学, 2021, 48(7): 256-269. https://doi.org/10.11896/jsjkx.200800223
[14] 暴雨轩, 芦天亮, 杜彦辉, 石达.
基于i_ResNet34模型和数据增强的深度伪造视频检测方法
Deepfake Videos Detection Method Based on i_ResNet34 Model and Data Augmentation
计算机科学, 2021, 48(7): 77-85. https://doi.org/10.11896/jsjkx.210300258
[15] 曾伟良, 陈漪皓, 姚若愚, 廖睿翔, 孙为军.
时空图注意力网络在交叉口车辆轨迹预测的应用
Application of Spatial-Temporal Graph Attention Networks in Trajectory Prediction for Vehicles at Intersections
计算机科学, 2021, 48(6A): 334-341. https://doi.org/10.11896/jsjkx.200800066
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!