{"type":"ScholarlyArticle","title":"Mutual Distillation Learning Network for Trajectory-User Linking","authors":[{"name":"Wei Chen","slug":"wei-chen","given":"Wei","family":"Chen","affiliation":"College of Computer Science and Technology , Ocean University of China"},{"name":"ShuZhe Li","slug":"shuzhe-li","given":"ShuZhe","family":"Li","affiliation":"College of Computer Science and Technology , Ocean University of China"},{"name":"Chao Huang","slug":"chao-huang","given":"Chao","family":"Huang","affiliation":"Department of Computer Science , The University of Hong Kong"},{"name":"Yanwei Yu","slug":"yanwei-yu","given":"Yanwei","family":"Yu","affiliation":"College of Computer Science and Technology , Ocean University of China"},{"name":"Yongguo Jiang","slug":"yongguo-jiang","given":"Yongguo","family":"Jiang","affiliation":"College of Computer Science and Technology , Ocean University of China"},{"name":"Junyu Dong","slug":"junyu-dong","given":"Junyu","family":"Dong","affiliation":"College of Computer Science and Technology , Ocean University of China"}],"abstract":"Trajectory-User Linking (TUL), which links trajectories to users who generate them, has been a challenging problem due to the sparsity in check-in mobility data. Existing methods ignore the utilization of historical data or rich contextual features in check-in data, resulting in poor performance for TUL task. \r\nIn this paper, we propose a novel Mutual distillation learning network to solve the TUL problem for sparse check-in mobility data, named MainTUL.  \r\nSpecifically, MainTUL is composed of a Recurrent Neural Network (RNN) trajectory encoder that models sequential patterns of input trajectory and a temporal-aware Transformer trajectory encoder that captures long-term time dependencies for the corresponding augmented historical trajectories. \r\nThen, the knowledge learned on historical trajectories is transferred between the two trajectory encoders to guide the learning of both encoders to achieve mutual distillation of information. Experimental results on two real-world check-in mobility datasets demonstrate the superiority of \\model against state-of-the-art baselines. The source code of our model is available at https://github.com/Onedean/MainTUL.","doi":"10.24963/ijcai.2022/274","url":"https://ijcai.org/proceedings/2022/274","pdf":"https://ijcai.org/proceedings/2022/0274.pdf","markdown":"https://ijcai.org/proceedings/2022/274.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/274","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"Data Mining","track":"Main Track","number":274,"pages":{"first":1973,"last":1979},"keywords":["Machine Learning: Semi-Supervised Learning","Machine Learning: Representation learning","Data Mining: Mining Spatial and/or Temporal Data"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38987071","length_sec":87,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38985214","length_sec":623,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Dai et al., 2021] Shaojie Dai, Jinshuai Wang, Chao Huang, et al. Temporal multi-view graph convolutional networks for citywide traffic volume inference. In ICDM, pages 1042-1047. IEEE, 2021.","title":"Temporal multi-view graph convolutional networks for citywide traffic volume inference","authors":"Dai","year":"2021","venue":"ICDM","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Gao et al., 2017] Qiang Gao, Fan Zhou, Kunpeng Zhang, et al. Identifying human mobility via trajectory embed- dings. In IJCAI, volume 17, pages 1689-1695, 2017.","title":"Identifying human mobility via trajectory embeddings","authors":"Qiang Gao, Fan Zhou, Kunpeng Zhang","year":"2017","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2017/234"},{"position":3,"raw":"[Hao et al., 2020] Qianyue Hao, Lin Chen, Fengli Xu, and Yong Li. Understanding the urban pandemic spreading of covid-19 with real world mobility data. In KDD, pages 3485-3492, 2020.","title":"Understanding the urban pandemic spreading of covid-19 with real world mobility data","authors":"Qianyue Hao, Lin Chen, Fengli Xu, Yong Li","year":"2020","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Hinton et al., 2015] Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531, 2015.","title":"Distilling the knowledge in a neural network","authors":"Geoffrey Hinton, Oriol Vinyals, Jeff Dean","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Hochreiter and Schmidhuber, 1997] Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural computation, 9(8):1735-1780, 1997.","title":"Long short-term memory","authors":"Schmidhuber Hochreiter, Sepp Hochreiter, Jürgen Schmidhuber","year":"1997","venue":"Neural computation","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Huang et al., 2018] Chao Huang, Junbo Zhang, et al. Deep- crime: Attentive hierarchical recurrent networks for crime prediction. In CIKM, pages 1423-1432, 2018.","title":"Deepcrime: Attentive hierarchical recurrent networks for crime prediction","authors":"Chao Huang, Junbo Zhang","year":"2018","venue":"CIKM","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Huang et al., 2019] Chao Huang, Chuxu Zhang, Jiashu Zhao, et al. Mist: A multiview and multimodal spatial- temporal learning framework for citywide abnormal event forecasting. In WWW, 2019.","title":"Mist: A multiview and multimodal spatialtemporal learning framework for citywide abnormal event forecasting","authors":"Chao Huang, Chuxu Zhang, Jiashu Zhao","year":"2019","venue":"WWW","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Jiang, 2018] Zhe Jiang. A survey on spatial prediction methods. TKDE, 31(9):1645-1664, 2018.","title":"A survey on spatial prediction methods","authors":"Zhe Jiang","year":"2018","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Jin et al., 2019] Fengmei Jin, Wen Hua, et al. Moving ob- ject linking based on historical trace. In ICDE, pages 1058-1069. IEEE, 2019.","title":"Moving object linking based on historical trace","authors":"Jin","year":"2019","venue":"ICDE","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Jin et al., 2020] Fengmei Jin, Wen Hua, et al. Trajectory- based spatiotemporal entity linking. TKDE, 2020.","title":"Trajectorybased spatiotemporal entity linking","authors":"Jin","year":"2020","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Keogh and Pazzani, 2000] Eamonn J Keogh and Michael J Pazzani. Scaling up dynamic time warping for datamining applications. In KDD, pages 285-289, 2000.","title":"Scaling up dynamic time warping for datamining applications","authors":"J Eamonn, Michael J Keogh, Pazzani","year":"2000","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Li et al., 2018] Xiucheng Li, Kaiqi Zhao, Gao Cong, et al. Deep representation learning for trajectory similarity com- putation. In ICDE, pages 617-628. IEEE, 2018.","title":"Deep representation learning for trajectory similarity computation","authors":"Li","year":"2018","venue":"ICDE","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Lian et al., 2014] Defu Lian, Yin Zhu, Xing Xie, et al. Ana- lyzing location predictability on location-based social net- works. In PAKDD, pages 102-113. Springer, 2014.","title":"Analyzing location predictability on location-based social networks","authors":"Defu Lian, Yin Zhu, Xing Xie","year":"2014","venue":"PAKDD","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Lin et al., 2020] Yan Lin, Huaiyu Wan, et al. Pre-training context and time aware location embeddings from spatial- temporal trajectories for user next location prediction. In AAAI, 2020.","title":"Pre-training context and time aware location embeddings from spatialtemporal trajectories for user next location prediction","authors":"Yan Lin, Huaiyu Wan","year":"2020","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Liu et al., 2014] Yong Liu, Wei Wei, Aixin Sun, et al. Ex- ploiting geographical neighborhood characteristics for lo- cation recommendation. In CIKM, pages 739-748, 2014.","title":"Exploiting geographical neighborhood characteristics for location recommendation","authors":"Yong Liu, Wei Wei, Aixin Sun","year":"2014","venue":"CIKM","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Miao et al., 2020] Congcong Miao, Jilong Wang, Heng Yu, et al. Trajectory-user linking with attentive recurrent net- work. In AAMAS, pages 878-886, 2020.","title":"Trajectory-user linking with attentive recurrent network","authors":"Miao","year":"2020","venue":"AAMAS","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Ruffy and Chahal, 2019] Fabian Ruffy and Karanbir Cha- hal. The state of knowledge distillation for classification. arXiv preprint arXiv:1912.10850, 2019.","title":"Fabian Ruffy and Karanbir Chahal. The state of knowledge distillation for classification","authors":"Chahal Ruffy","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Shahdoosti and Mirzapour, 2017] Hamid Reza Shahdoosti and Fardin Mirzapour. Spectral-spatial feature extraction using orthogonal linear discriminant analysis for classifi- cation of hyperspectral data. European Journal of Remote Sensing, 50(1):111-124, 2017.","title":"Hamid Reza Shahdoosti and Fardin Mirzapour. Spectral-spatial feature extraction using orthogonal linear discriminant analysis for classification of hyperspectral data","authors":"Mirzapour Shahdoosti","year":"2017","venue":"European Journal of Remote Sensing","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Shang et al., 2017] Shuo Shang, Lisi Chen, Zhewei Wei, et al. Trajectory similarity join in spatial networks. VLDB, 10(11), 2017.","title":"Trajectory similarity join in spatial networks","authors":"Shuo Shang, Lisi Chen, Zhewei Wei","year":"2017","venue":"VLDB","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Vaswani et al., 2017] Ashish Vaswani, Noam Shazeer, Niki Parmar, et al. Attention is all you need. In NeurIPS, pages 5998-6008, 2017.","title":"Attention is all you need","authors":"Ashish Vaswani, Noam Shazeer, Niki Parmar","year":"2017","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Wu et al., 2020] Xian Wu, Chao Huang, Chuxu Zhang, et al. Hierarchically structured transformer networks for fine-grained spatial event forecasting. In WWW, pages 2320-2330, 2020.","title":"Hierarchically structured transformer networks for fine-grained spatial event forecasting","authors":"Xian Wu, Chao Huang, Chuxu Zhang","year":"2020","venue":"WWW","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Yang et al., 2015] Dingqi Yang, Daqing Zhang, et al. Mod- eling user activity preference by leveraging user spatial temporal characteristics in lbsns. Transactions On SMC, 45(1):129-142, 2015.","title":"Modeling user activity preference by leveraging user spatial temporal characteristics in lbsns","authors":"Yang","year":"2015","venue":"Transactions On SMC","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Yang et al., 2019] Zhilin Yang, Zihang Dai, Yiming Yang, et al. Xlnet: Generalized autoregressive pretraining for language understanding. NeurIPS, 32, 2019.","title":"Xlnet: Generalized autoregressive pretraining for language understanding","authors":"Zhilin Yang, Zihang Dai, Yiming Yang","year":"2019","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Yang et al., 2021] Peilun Yang, Hanchen Wang, Ying Zhang, et al. T3s: Effective representation learning for trajectory similarity computation. In ICDE, pages 2183- 2188. IEEE, 2021.","title":"T3s: Effective representation learning for trajectory similarity computation","authors":"Yang","year":"2021","venue":"ICDE","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Yao et al., 2020] Di Yao, Gao Cong, Chao Zhang, et al. A linear time approach to computing time series similarity based on deep metric learning. TKDE, 2020.","title":"A linear time approach to computing time series similarity based on deep metric learning","authors":"Di Yao, Gao Cong, Chao Zhang","year":"2020","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Ying et al., 2010] Josh Jia-Ching Ying, Eric Hsueh-Chan Lu, Wang-Chien Lee, et al. Mining user similarity from semantic trajectories. In SIGSPATIAL Workshop on LB- SNs, pages 19-26, 2010.","title":"Mining user similarity from semantic trajectories","authors":"Josh Jia-Ching Ying, Eric Hsueh-Chan, Wang-Chien Lu, Lee","year":"2010","venue":"SIGSPATIAL Workshop on LB-SNs","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhang et al., 2020] Hanyuan Zhang, Xinyu Zhang, Qize Jiang, et al. Trajectory similarity learning with auxiliary supervision and optimal matching. In IJCAI, pages 3209- 3215, 2020.","title":"Trajectory similarity learning with auxiliary supervision and optimal matching","authors":"Hanyuan Zhang, Xinyu Zhang, Qize Jiang","year":"2020","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2020/444"},{"position":28,"raw":"[Zhao et al., 2021] Jiawei Zhao, Wei Luo, Boxing Chen, et al. Mutual-learning improves end-to-end speech trans- lation. In EMNLP, pages 3989-3994, 2021.","title":"Mutual-learning improves end-to-end speech translation","authors":"Jiawei Zhao, Wei Luo, Boxing Chen","year":"2021","venue":"EMNLP","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zhou et al., 2018] Fan Zhou, Qiang Gao, Goce Trajcevski, et al. Trajectory-user linking via variational autoencoder. In IJCAI, pages 3212-3218, 2018.","title":"Trajectory-user linking via variational autoencoder","authors":"Fan Zhou, Qiang Gao, Goce Trajcevski","year":"2018","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2018/446"},{"position":30,"raw":"[Zhou et al., 2021a] Fan Zhou, Yurou Dai, Qiang Gao, et al. Self-supervised human mobility learning for next location prediction and trajectory classification. KBS, page 107214, 2021.","title":"Self-supervised human mobility learning for next location prediction and trajectory classification","authors":"Zhou","year":"2021","venue":"KBS","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Zhou et al., 2021b] Fan Zhou, Ruiyang Yin, et al. Improv- ing human mobility identification with trajectory augmen- tation. GeoInformatica, 25(3):453-483, 2021.","title":"Improving human mobility identification with trajectory augmentation","authors":"Zhou","year":"2021","venue":"GeoInformatica","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Zuo et al., 2020] Simiao Zuo, Haoming Jiang, Zichong Li, et al. Transformer hawkes process. In ICML, pages 11692-11702. PMLR, 2020.","title":"Transformer hawkes process","authors":"Zuo","year":"2020","venue":"ICML","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}