TY - JOUR AU - Liu, Zhicheng AU - Miranda, Fabio AU - Xiong, Weiting AU - Yang, Junyan AU - Wang, Qiao AU - Silva, Claudio PY - 2020/04/03 Y2 - 2024/03/28 TI - Learning Geo-Contextual Embeddings for Commuting Flow Prediction JF - Proceedings of the AAAI Conference on Artificial Intelligence JA - AAAI VL - 34 IS - 01 SE - AAAI Technical Track: Applications DO - 10.1609/aaai.v34i01.5425 UR - https://ojs.aaai.org/index.php/AAAI/article/view/5425 SP - 808-816 AB - <p>Predicting commuting flows based on infrastructure and land-use information is critical for urban planning and public policy development. However, it is a challenging task given the complex patterns of commuting flows. Conventional models, such as gravity model, are mainly derived from physics principles and limited by their predictive power in real-world scenarios where many factors need to be considered. Meanwhile, most existing machine learning-based methods ignore the spatial correlations and fail to model the influence of nearby regions. To address these issues, we propose <strong><span style="text-decoration: underline;">G</strong></span>eo-contextual <strong><span style="text-decoration: underline;">M</strong></span>ultitask <strong><span style="text-decoration: underline;">E</strong></span>mbedding <strong><span style="text-decoration: underline;">L</strong></span>earner (<strong>GMEL</strong>), a model that captures the spatial correlations from geographic contextual information for commuting flow prediction. Specifically, we first construct a geo-adjacency network containing the geographic contextual information. Then, an attention mechanism is proposed based on the framework of graph attention network (GAT) to capture the spatial correlations and encode geographic contextual information to embedding space. Two separate GATs are used to model supply and demand characteristics. To enhance the effectiveness of the embedding representation, a multitask learning framework is used to introduce stronger restrictions, forcing the embeddings to encapsulate effective representation for flow prediction. Finally, a gradient boosting machine is trained based on the learned embeddings to predict commuting flows. We evaluate our model using real-world dataset from New York City and the experimental results demonstrate the effectiveness of our proposed method against the state of the art.</p> ER -