Introduction to the Special Issue on Machine Learning and Location Data: Part II

Introduction to the Special Issue on Machine Learning and Location Data: Part IIBig location data led to the proposal of various spatial algorithms, models, and systems to understand and model these data for intelligent transportation, business intelligence, public health, social economics, urban planning, urban resilience, and environmental sustainability.The second part of the ACM Transactions on Spatial Algorithms and Systems (TSAS) special issue on Machine Learning and Location Data includes seven innovative articles that address key challenges in areas such as trajectory re-identification, geospatial representation learning, traffic prediction, and spatial query optimization.The first two articles focus on advancements in the representation learning of GPS trajectories and geospatial entities.The first article, titled "RE-Trace: Re-Identification of Modified GPS Trajectories, " presents a contrastive learning model that facilitates reliable and efficient re-identification of GPS trajectories and resists specific trajectory transformation attacks aimed to obscure a trajectory's origin.The second article, titled "Latent Representation Learning for Geospatial Entities, " introduces GeoNN, a novel graph neural network-based model designed to learn spatially aware embeddings for geospatial entities.The next three articles center on forecasting technologies for traffic flow and multivariate time series.The third article, titled "Cross-and Context-aware Attention-based Spatial-temporal Graph Convolutional Networks for Human Mobility Prediction, " incorporates cross-and context-aware attention mechanisms to capture both spatial and temporal dependencies in traffic flow patterns.The fourth article, titled "Overcoming Catastrophic Forgetting in Continual Fine-grained Urban Flow Inference, " is an extended version of a conference paper presented at AAAI 2023, proposing a novel framework for inferring fine-grained citywide traffic flows.The fifth article, titled "Dynamic Spatial-temporal Embedding via Neural Conditional Random Field for Multivariate Time Series Forecasting, " introduces two significant plugins to help automatically learn enhanced dynamic spatialtemporal embeddings of multivariate time series data.The final two articles focus on query optimization for spatial databases.The sixth article, titled "A Generic Machine Learning Model for Spatial Query Optimization Based on Spatial Embeddings, " proposes a new approach that leverages spatial embeddings to address the computational challenges of querying large spatial datasets.The seventh article, titled "Backbone Index and GNN Models for Skyline Path Query Evaluation over Multi-cost Road Networks, " presents a novel index structure along with a corresponding index construction method that condenses a multi-cost road network to multiple smaller summarized graphs with different granularity.The diverse approaches presented in these articles not only contribute to the foundational methodologies of machine learning and location data but also open new pathways for further research and practical applications in location-based services.We extend our heartfelt thanks to the authors for their contributions and to the reviewers for their insightful and timely feedback, ensuring the high quality of the accepted articles.