Predicting the Next Location: A Recurrent Model with Spatial and Temporal Contexts

By Qiang Liu, Shu Wu, Liang Wang, Tien-Ping Tan. 2016

1,026 citations recorded by the source corpus.

Spatial and temporal contextual information plays a key role for analyzing user behaviors, and is helpful for predicting where he or she will go next. With the growing ability of collecting information, more and more temporal and spatial contextual information is collected in systems, and the location prediction problem becomes crucial and feasible. Some works have been proposed to address this problem, but they all have their limitations. Factorizing Personalized Markov Chain (FPMC) is constructed based on a strong independence assumption among different factors, which limits its performance. Tensor Factorization (TF) faces the cold start problem in predicting future actions. Recurrent Neural Networks (RNN) model shows promising performance comparing with PFMC and TF, but all these methods have problem in modeling continuous time interval and geographical distance. In this paper, we extend RNN and propose a novel method called Spatial Temporal Recurrent Neural Networks (ST-RNN). ST-RNN can model local temporal and spatial contexts in each layer with time-specific transition matrices for different time intervals and distance-specific transition matrices for different geographical distances. Experimental results show that the proposed ST-RNN model yields significant improvements over the competitive compared methods on two typical datasets, i.e., Global Terrorism Database (GTD) and Gowalla dataset.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Transportation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Information Systems

In Transportation

Citations in the corpus

A Survey on Trajectory Data Management, Analytics, and Learning (2020, methodology): “ly used for trajectory prediction, and most of the work reviewed also learns user representations from trajectories to capture user preferences in addition to the spatio-temporal contexts. Liu et al. [113] proposed Spatial-Temporal RNN (ST-RNN) for the next location prediction of a given user using ”
STAN: Spatio-Temporal Attention Network for Next Location Recommendation (2021, methodology): “• STRNN [22]: an invariant RNN model that incorporates spatio-temporal features between consecutive visits.”
Sequence-Aware Recommender Systems (2018, background): “ et al. [56] SA I EC Jannach et al. [57] SA WI MUS Jannach and Ludewig [54] SA — EC Lerche et al. [70] SA I EC Letham et al. [71] LI I EC Lian et al. [72] LI I POI Lim et al. [73] LI S POI Liu et al. [76] LI SI POI Liu et al. [77] LI I POI Lu et al. [78] SB I APP Maillet et al. [80] — W MUS McFee an”
DeepMove: Predicting Human Mobility with Attentional Recurrent Networks (2018, methodology): “We adapt the ST-RNN [20] to our scene where only anonymous location ID is known, while the original version focuses on modeling the continuous spatiotemporal information.”
Deep Learning for Spatio-Temporal Data Mining: A Survey (2019, neutral): “Application domains Related works Transportation [177], [187], [116], [31], [29], [6], [32], [127], [53], [112], [26], [105], [166], [82], [23], [66], [108], [126], [81], [139], [78], [128], [146], [148], [72], [98], [189], [22], [24], [99], [96], [16], [97], [85], [15], [164], [165], [95], [140], [”
Where to Go Next: A Spatio-Temporal Gated Network for Next POI Recommendation (2019, methodology): “• ST-RNN (Liu et al. 2016a): Based on the standard RNN model, ST-RNN replaces the single transition matrix in RNN with time-speciﬁc transition matrices and distance-speciﬁc transition matrices to model spatial and temporal contexts.”
A Survey on Deep Learning for Human Mobility (2020, methodology): “ST-RNN (Spatial Temporal Recurrent Neural Networks) [114] overtakes this issue by extending RNN with timeand spatial-specific transition matrices.”
SS-LSTM: A Hierarchical LSTM Model for Pedestrian Trajectory Prediction (2018, methodology): “Recently, LSTM-based methods have been used in citywide level applications like public transportation prediction [21] and location prediction problem [17].”
Where to Go Next: Modeling Long- and Short-Term User Preferences for Point-of-Interest Recommendation (2020, methodology): “Inspired by the success of RNN in sequential data modeling (Hidasi et al. 2015; Chen et al. 2019a; 2018; Huang et al. 2018), RNN based methods become pervasive in the ﬁeld of next-POI recommendation (Liu et al. 2016; Manotumruksa, Macdonald, and Ounis.”