Among the diverse services provided by Location-Based Social Networks (LBSNs), Next Point-of-Interest (POI) recommendation plays a crucial role in inferring user preferences from historical check-in trajectories. However, existing sequential and graph-based methods frequently neglect significant mobility variations across distinct contextual scenarios (e.g., tourists versus locals). This oversight results in suboptimal performance due to two fundamental limitations: the inability to capture scenario-specific features and the failure to resolve inherent inter-scenario conflicts. To overcome these limitations, we propose the Multifaceted Scenario-Aware Hypergraph Learning method (MSAHG), a framework that adopts a scenario-splitting paradigm for next POI recommendation. Our main contributions are: (1) Construction of scenario-specific, multi-view disentangled sub-hypergraphs to capture distinct mobility patterns; (2) A parameter-splitting mechanism to adaptively resolve conflicting optimization directions across scenarios while preserving generalization capability. Extensive experiments on three real-world datasets demonstrate that MSAHG consistently outperforms five state-of-the-art methods across diverse scenarios, confirming its effectiveness in multi-scenario POI recommendation.

Semantic segmentation under adverse imaging conditions such as low illumination and motion blur often breaks into holes, fragments, and disconnected regions. Naive smoothing can reduce these artifacts, but it also tends to leak labels across true boundaries. We address this coherence–leakage dilemma with a plug-and-play decode-head refinement composed of aMulti-scale Boundary Detector (MBD)and aBoundary-Barrier Coherence Refiner (BBCR). MBD predicts a boundary probability map, and BBCR turns it into a soft barrier for logit-space refinement through prototype-based global coherence and barrier-gated local affinity propagation. Targeting mixed-condition deployments, where favorable and adverse inputs are interleaved, the proposed modules attach to existing segmentation heads without modifying backbones or necks. Across eight architectures on a binary dairy-farm dataset, our method improves mIoU by 0.97 points on average and by up to 13.35 points on a manually annotated low-light/blur stress-test split without target-domain fine-tuning. On NightCity with 19 classes, it further yields a consistent average gain of 0.96 mIoU points while preserving performance under favorable conditions.

Trajectory anomaly detection is critical for transportation safety, solving traffic congestion, optimizing routes, and other smart city applications. Existing methods often misclassify reasonable but non-standard trajectories, such as route changes due to traffic congestion, as anomalies. Additionally, many methods require extensive labeled datasets for training, however, in real-world scenarios, trajectory anomaly labeled data are often scarce and costly to obtain. To address the challenges, we investigate a Semi-Supervised Contrastive learning approach for Trajectory Anomaly Detection, namely SSCTAD. Specifically, SSCTAD integrates a transformer based method to process trajectories and utilizes the in-out flow encoder to capture the spatial-temporal patterns from the trajectories, which can help the model capture the traffic conditions. Then, by leveraging a small subset of labeled data along with extensive unlabeled data, SSCTAD utilizes a semi-supervised contrastive learning component enables the model to better distinguish normal and anomalous trajectories by maximizing the similarity of latent representation from the same class and diversifying those from different classes. This method improves the performance of trajectory anomaly detection and effectively reduces the dependence of the model on the labeled dataset. Finally, experiments show that SSCTAD outperforms the other baseline models.

The development of the Internet of Things (IoT) has enhanced smart city services for traffic monitoring, leading to numerous schemes for accurate flow prediction based on traffic sensors. However, existing approaches primarily capture spatio-temporal (ST) dependencies from traffic graphs and train their models using randomly ordered data. This overlooks the fact that the modeling difficulty of each sensor/node in the ST traffic graph can vary significantly due to its spatial dependencies and temporal trends, resulting in unreliable and unstable predictions in IoT scenarios. In this context, we argue that a well-designed curriculum with an easy-to-difficult order can improve the training of ST models. Therefore, this paper introduces an ST difficulty measurer to score the node-level difficulty of traffic graph from both spatial and temporal aspects, and then implements a curriculum in the ST model training process. More specifically, based on the tiered ST difficulty score, the ST model training begins with a subgraph consisting of “easy” nodes characterized by relatively consistent spatial relationships and regular temporal patterns. Gradually, more difficult nodes are incorporated into the subgraph and participate in subsequent training stages. Comprehensive experiments and analysis on two real-world traffic flow datasets confirm the effectiveness of our proposed approach.

Maritime traffic analysis plays a vital role in port operations and logistics coordination. Although Automatic Identification System (AIS) provides rich vessel movement data, analyzing port-to-port traffic remains challenging due to data heterogeneity and missing trajectory segments. We present PortVIS, a web-based interactive system that supports the comprehensive analysis and visualization of maritime traffic. PortVIS integrates multi-sourced datasets—including AIS records, port and anchorage metadata, and wind field data—to enable trajectory segmentation, regional analysis, and data imputation. Users can import raw AIS records and segment them into port-to-port trips using our system. These segmented trajectories can then be filtered and queried based on vessel information and trip attributes. Users can also define custom zones (e.g., anchorages or transit areas) and explore traffic patterns through maps and charts. Missing trajectory segments are reconstructed using our recent imputation approach to improve data quality. By integrating trajectory processing and various visual analytics, PortVIS provides a unified tool for maritime mobility analysis.

Large-scale human mobility exhibits spatial and temporal patterns that can assist policymakers in decision making. Although traditional prediction models attempt to capture these patterns, they are often affected by nonperiodic public events, such as disasters and occasional celebrations. Since regular human mobility patterns are affected by these events, estimating their causal effects is critical to accurate mobility predictions. News articles provide unique perspectives on these events, though processing them is a challenge. In this study, we propose a causality based prediction model, CausalMob, to analyze the causal effects of public events. We first utilize large language models (LLMs) to extract human intentions from news and transform them into features that act as causal treatments. Next, the model learns representations of spatio-temporal regional covariates from multiple data sources to serve as confounders for causal inference. Finally, we present a causal effect estimation framework to ensure that event features remain independent of confounders during prediction. Based on large-scale real-world data, the experimental results show that the proposed model excels in human mobility prediction, outperforming state-of-the-art models.

Recent advancements have suggested that neural radiance fields (NeRFs) show great potential in color editing within the 3D domain. However, most existing NeRF-based editing methods continue to face significant challenges in local region editing, which usually lead to imprecise local object boundaries, difficulties in maintaining multi-view consistency, and over-reliance on annotated data. To address these limitations, in this paper, we propose a novel weakly-supervised method called KaRF for local color editing, which facilitates high-fidelity and realistic appearance edits in arbitrary regions of 3D scenes. At the core of the proposed KaRF approach is a unified two-stage Kolmogorov-Arnold Networks (KANs)-based radiance fields framework, comprising a segmentation stage followed by a local recoloring stage. This architecture seamlessly integrates geometric priors from NeRF to achieve weakly-supervised learning, leading to superior performance. More specifically, we propose a residual adaptive gating KAN structure, which integrates KAN with residual connections, adaptive parameters, and gating mechanisms to effectively enhance segmentation accuracy and refine specific editing effects. Additionally, we propose a palette-adaptive reconstruction loss, which can enhance the accuracy of additive mixing results. Extensive experiments demonstrate that the proposed KaRF algorithm significantly outperforms many state-of-the-art methods both qualitatively and quantitatively. Our code and more results are available at: https://github.com/PaiDii/KARF.git.

This paper presents a simulation for traffic evacuation during railway disruptions to enhance urban resilience. The research focuses on large-scale railway networks and provides flexible simulation settings to accommodate multiple node or line failures. The evacuation optimization model is mathematically formulated using matrix computation and nonlinear programming. The simulation integrates railway lines operated by various companies, along with external geographical features of the network. Furthermore, to address computational complexity in large-scale graph networks, a subgraph partitioning solution is employed for computation acceleration. The model is evaluated using the extensive railway network of Greater Tokyo. Data collection included both railway network structure and real-world GPS footfall data to estimate the number of station-area visitors for simulation input and evaluation purposes. Several evacuation scenarios were simulated for major stations including Tokyo, Shinjuku, Shibuya and so on. The results demonstrate that both evacuation passenger flow (EPF) and average travel time (ATT) during emergencies were successfully optimized, while remaining within the capacity constraints of neighboring stations and the targeted disruption recovery times.

Traffic states typically exhibit regular spatio-temporal patterns that support accurate prediction. However, traffic accidents can disrupt these patterns and introduce irregular fluctuations that challenge forecasting models. Previous studies have focused on predicting accident hotspots, risks, or traffic states using simple accident features. Yet, because traffic accidents are infrequent and exhibit spatio-temporal biases relative to collected traffic sensor data, they may significantly or only slightly affect traffic states, posing challenges for accurate prediction. In this study, we propose a large language model (LLM) enhanced Causality-aware Traffic accidents post-effects prediction (LLeCaT) to uncover the causal relationship between traffic states and accidents. Our model integrates fine-tuned LLMs to extract semantic information from accident records and estimate their effects on future traffic states following their occurrence. To validate the effectiveness of our proposed method, we collected and processed publicly available anonymized traffic accident records, traffic state datasets, and environmental information. We conducted extensive experiments demonstrating its superiority over all baselines. Additional ablation studies and case analyses further confirm the validity of the estimated causal effects.

Friend recommendation is an important real-world application in Location-based Social Networks (LBSN), helping users discover potential friends and enhance their overall happiness. LBSN mainly comprises two distinct data structures: spatio-temporal data for human mobility and graph data for social networks. These two data structures make it challenging to model the complex relationships between them, which are essential for comprehensively understanding users’ lives. Previous studies have either modeled user trajectories and social networks separately or used classical simple graph-based methods, where a simple edge links only two nodes, failing to capture the multiple relationships inherent in LBSN. Furthermore, most studies have relied on Euclidean space to train their graph models, which could result in significant distortion because of tree-like social network data structure. To address these limitations, we propose a novel heterogeneous LBSN hypergraph that represents user check-in records and continuous trajectories—comprising multiple Points of Interest (POI)—as hyperedges, enabling the representation of complex spatio-temporal relationships. This approach enables us to link multiple nodes of different types by hyperedges and use hyperbolic spaces to create more efficient graph representations. Additionally, we devise a new type-specific attention mechanism for our Heterogeneous Hyperbolic Hypergraph Neural Network (H 3 GNN), which is end-to-end trainable and employs supervised contrastive learning to learn hypergraph node embeddings for the subsequent friend recommendation task with the help of hyperbolic space. Finally, our model H 3 GNN achieves better results than existing methods on six real-world city datasets, and our ablation studies demonstrate the effectiveness of each component. Additionally, our experiments indicate that H 3 GNN requires less data storage and training time compared to previous methods.

Analyzing the causal impact of various government-related policies on the epidemic spread is of critical importance. This article aims to investigate the problem of assessing the causal effects of different COVID-19-related policies on the USA epidemic spread in different counties at any given time period, while eliminating biased interference from unobserved confounders (e.g., the vigilance of residents). However, the infection outcome of each region is influenced not only by its own confounding factors but also by policy interventions implemented in neighboring regions. Furthermore, the government policy index may exhibit a time-delay influence on outbreak dynamics. To this end, we implement observational data about different COVID-19-related policies (treatment) and outbreak dynamics (outcome) across different U.S. counties over time and develop a causal framework that learns the representations of time-varying confounders to tackle the aforementioned issues. More specifically, we employ one recurrent structure to capture the accumulative effects stemming from the policy history and then utilize hypergraph neural network to model the interactions among spatial regions. Our experimental results demonstrate the effectiveness of the proposed framework in quantifying the causal impact of different policy types on epidemics. Compared with baseline methods, our assessment provides valuable insights for future policy-making endeavors.

Maritime transportation, vital for nearly 90% of global trade, necessitates precise vessel trajectory prediction for safety and efficiency. Although the Automatic Identification System (AIS) provides a comprehensive data source, how to model these multi-modal and heterogeneous time-varying sequences (such as vessels’ kinetic information and ocean weather factors) poses a formidable challenge. Moreover, most existing approaches are limited by the confined scope of vessel trajectory modeling, making it impossible to consider the unique characteristics of maritime transportation system. To tackle these challenges, we propose a novel framework called AISFuser to i) encode unique maritime traffic network into graphical representations, and ii) introduce the heterogeneity into multi-modal temporal embeddings through Self-Supervised Learning (SSL). Specifically, our AISFuser is constructed by combining an attention-based graph block with a transformer network to encode information across space and time, respectively. In terms of temporal dimension, one SSL auxiliary task is also designed to enhance the heterogeneity of temporal representations and supplement the main vessel prediction task. We validate the effectiveness of the proposed AISFuser on a real-world AIS dataset. Extensive experimental results demonstrate that our method can forecast multiple attributes of vessel trajectory for over 10 hours into the future, outperforming competitive baselines.

In many intelligent transportation systems, predicting the future motion of heterogeneous traffic participants is a fundamental but challenging task due to various factors encompassing the agents’ dynamic states, interactions with neighboring agents and surrounding traffic infrastructures, and their stochastic and multi-modal natural behavior tendencies. However, existing approaches have limitations as they either focus solely on static, pairwise interactions, ignoring interactions of varied granularity, or fail to tackle agents’ heterogeneity. In this paper, instead of focusing solely on pairwise interactions, we propose a Heterogenous Hypergraph Graph Neural Network (HHGNN) based motion prediction model that leverages the nature of hypergraph to encode the groupwise interactions among traffic participants. Moreover, we propose the type-aware two-level hypergraph message passing module (TTHMS) with learnable hyperedge-type embeddings to model the intra-group and inter-group level interactions among heterogeneous traffic agents (e.g., vehicles, pedestrians, and cyclists). Besides, We integrate a scene context fusion layer in TTHMS to incorporate the scene context. Comparison and ablation experiments on the Waymo Open Motion Dataset (WOMD) demonstrate HHGNN’s effectiveness within the motion prediction task.

Maritime traffic management increasingly relies on vessel position information provided by terrestrial and satellite networks of the Automatic Identification System (AIS). Unfortunately, the problem of missing AIS data can lead to long-term gaps in vessel trajectory, raising corresponding security concerns regarding collision risks and illicit activities. Existing imputation approaches are often constrained by vehicle-based low-sampling trajectories, hindering their ability to address unique characteristics of maritime transportation systems and long-term missing scenarios. To tackle these challenges, we propose a novel generative framework for long-term vessel trajectory imputation. Our framework considers irregular tracks of vessels, which differ from those of cars due to the absence of a structured road network, and ensures the continuity of multi-point imputed trajectories. Specifically, we first utilize a pre-trained trajectory embedding block to capture patterns of vessel movements. Subsequently, we introduce a diffusion-based model for generating missing trajectories, where observed trajectory modeling with transformer encoding architecture and embeddings of both historical vessel trajectory and external factors serve as conditional information. In particular, we design a physics-guided discriminator in the training stage, which imposes kinematic constraints between locations and angles to improve the continuity of the imputed trajectories. Comprehensive experiments and analysis on a real-world AIS dataset confirm the effectiveness of our proposed approach.

With the popularity of location-based services, human mobility prediction plays a key role in enhancing personalized navigation, optimizing recommendation systems, and facilitating urban mobility and planning. This involves predicting a user's next POI (point-of-interest) visit using their past visit history. However, the uneven distribution of visitations over time and space, namely the long-tail problem in spatial distribution, makes it difficult for AI models to predict those POIs that are less visited by humans. In light of this issue, we propose the $\underline{\bf{Lo}}$ng-$\underline{\bf{T}}$ail Adjusted $\underline{\bf{Next}}$ POI Prediction (LoTNext) framework for mobility prediction, combining a Long-Tailed Graph Adjustment module to reduce the impact of the long-tailed nodes in the user-POI interaction graph and a novel Long-Tailed Loss Adjustment module to adjust loss by logit score and sample weight adjustment strategy. Also, we employ the auxiliary prediction task to enhance generalization and accuracy. Our experiments with two real-world trajectory datasets demonstrate that LoTNext significantly surpasses existing state-of-the-art works.

Recent achievements in deep learning (DL) have demonstrated its potential in predicting traffic flows. Such predictions are beneficial for understanding the situation and making traffic control decisions. However, most state-of-the-art DL models are considered “black boxes” with little to no transparency of the underlying mechanisms for end users. Some previous studies attempted to “open the black box” and increase the interpretability of generated predictions. However, handling complex models on large-scale spatiotemporal data and discovering salient spatial and temporal patterns that significantly influence traffic flow remain challenging. To overcome these challenges, we present TrafPS , a visual analytics approach for interpreting traffic prediction outcomes to support decision-making in traffic management and urban planning. The measurements region SHAP and trajectory SHAP are proposed to quantify the impact of flow patterns on urban traffic at different levels. Based on the task requirements from domain experts, we employed an interactive visual interface for the multi-aspect exploration and analysis of significant flow patterns. Two real-world case studies demonstrate the effectiveness of TrafPS in identifying key routes and providing decision-making support for urban planning.

Perceiving and modeling urban crowd movements are of great importance to smart city-related fields. Governments and public service operators can benefit from such efforts as they can be applied to crowd management, resource scheduling, and early emergency warning. However, most prior research on urban crowd modeling has failed to describe the dynamics and continuity of human mobility, leading to inconsistent and irrelevant results when they tackle multiple homogeneous forecasting tasks as they can only be modeled independently. To overcome this drawback, we propose to model human mobility from a new perspective, which uses the citywide crowd transition process constituted by a series of transition matrices from low order to high order, to help us understand how the crowd dynamics evolve step-by-step. We further propose a Deep Transition Process Network to process and predict such new high-dimensional data, where novel grid embedding with Graph Convolutional Network, parameter-shared Convolutional LSTM, and High-Dimensional Attention mechanism are designed to learn the complicated dependencies in terms of spatial, temporal, and ordinal features. We conduct experiments on two datasets generated by a large amount of GPS data collected from a real-world smartphone application. The experiment results demonstrate the superior performance of our proposed methodology over existing approaches.

Spatial-temporal (ST) graph modeling, such as traffic speed forecasting and taxi demand prediction, is an important task in deep learning area. However, for the nodes in the graph, their ST patterns can vary greatly in difficulties for modeling, owning to the heterogeneous nature of ST data. We argue that unveiling the nodes to the model in a meaningful order, from easy to complex, can provide performance improvements over traditional training procedure. The idea has its root in Curriculum Learning, which suggests in the early stage of training models can be sensitive to noise and difficult samples. In this paper, we propose ST-Curriculum Dropout, a novel and easy-to-implement strategy for spatial-temporal graph modeling. Specifically, we evaluate the learning difficulty of each node in high-level feature space and drop those difficult ones out to ensure the model only needs to handle fundamental ST relations at the beginning, before gradually moving to hard ones. Our strategy can be applied to any canonical deep learning architecture without extra trainable parameters, and extensive experiments on a wide range of datasets are conducted to illustrate that, by controlling the difficulty level of ST relations as the training progresses, the model is able to capture better representation of the data and thus yields better generalization.

To understand human mobility following the typhoon, analyzing the causal impact of extreme typhoon weather on human mobility is important for disaster emergency management. However, the unobserved confounders (e.g., the characteristic of each region) correlate with the strength of typhoon weather and also affect human mobility during typhoon, which may generate biased influences on the causal analysis process. Besides, these confounders may be time-varying following the dynamic movements of typhoon. In this work, we develop a neural network-based continuous causal effect estimation framework to mitigate the interference from (unobserved) confounders and assess the continuous causal responses of typhoon-related weather (treatment) on several types of human mobility (outcome) across different counties at any given period. To this end, we integrate the big data from two huge typhoons in Japan (i.e., Typhoon Faxai and Hagibis) and leverage multiple sources of covariates (i.e., residents' vigilance and basic mobility patterns) from different counties to learn the representations of time-varying confounders. The experimental results indicate the effectiveness of our proposed framework in capturing the confounders for quantifying the causal impact of extreme weather during the typhoon process, compared with several existing causal studies.

Machine unlearning allows data owners to erase certain data and its impact from learning models for the right to be forgotten. However, privacy risks during the unlearning process have been identified. Earlier studies have used differences in model outputs before and after unlearning to conduct membership inference attacks. Nevertheless, the current attacks on machine unlearning are limited to inference and cannot reconstruct data without access to the victim's dataset. In this paper, we propose a reconstruction attack towards machine unlearning (RAU), which can reconstruct the unlearned data by exploiting the privacy leakage from the two models. To improve reconstruction quality, we propose a Conditional Matching Generative Adversarial Network (CMGAN), a novel variant of generative adversarial networks which introduces a reconstructive loss. Our work demonstrates the possible privacy leakage of current machine unlearning scenarios. Experimental results on MNIST and Fashion-MNIST show that the proposed attack achieves high label recovery accuracy and good data recovery performance.

Building outline extracted from high-resolution aerial images can be used in various application fields such as change detection and disaster assessment. However, traditional CNN model cannot recognize contours very precisely from original images. In this paper, we proposed a CNN and Transformer based model together with active contour model to deal with this problem. We also designed a triple-branch decoder structure to handle different features generated by encoder. Experiment results show that our model outperforms other baseline model on two datasets, achieving 91.1% mIoU on Vaihingen and 83.8% on Bing huts.

This paper introduces a graph encoding-based hybrid vision transformer for automatic road network extraction via high-resolution remote sensing imagery. Given that high-resolution remote sensing images covered large urban areas, traditional segmentation-based road extraction methods usually can generate good binary classification maps in simple structured road surfaces but fail in complex highway and bridge-covered areas. We introduce a graph encoding-based mechanism to address the above issues, enabling the road extraction framework extracts the road segmentation feature and build the graph structure map jointly. Compared to only segmentation-based methods, our approach learns prior geometrical structure information from the extracted ViT feature maps and has a non-local awareness of the whole road network structure. Eexperimental results demonstrated that the proposed approach outperforms the traditional segmentation-based methods.

Today, thanks to the rapid developing mobile and sensor networks in IoT (Internet of Things) systems, spatio-temporal big data are being constantly generated. They have brought us a data-driven possibility to sense and understand crowd mobility on a city scale. A fundamental task towards the next-generation mobility services, such as Intelligent Transportation Systems (ITS), Mobility-as-a-Service (MaaS), is spatio-temporal predictive modeling of the geo-sensory signals. There is a recent line of research leveraging deep learning techniques to boost the forecasting performance on such tasks. While simulating the regularity of mobility behaviors (e.g., routines, periodicity) in a more sophisticated way, the existing studies ignore an important part of urban activities, i.e., events. Including holidays, extreme weathers, pandemic, accidents, various urban events happen from time to time and cause non-stationary phenomena, which by nature make the spatio-temporal forecasting task challenging. We thereby envision an event-aware urban mobility prediction model that is capable of fast adapting and making reliable predictions in different scenarios, which is crucial to decision making towards emergency response and urban resilience.

Mobility digital twin, which is a a virtual replica of the mobility in the physical world, is the key building block of modern smart city applications at a metropolitan scale, including traffic regulation, emergency management and epidemic control. To duplicate the mobility in the physical world and show the potential outcome based on either the current state or manipulated conditions, we are facing with three main challenges: 1) how to sense real-time human mobility at a large scale and assimilate different data sources to infer a dynamic city mobility state; 2) how to make an accurate prediction for the mobility replica that adapts with dynamic city mobility state; 3) how to simulate the mobility in response to different conditions. In this talk, we will present our recent studies, practices and perspectives on mobility digital twin in addressing these above challenges, with applications in the real-world scenarios with industrial connections.

In recent years, the use of WiFi fingerprints for indoor positioning has grown in popularity, largely due to the widespread availability of WiFi and the proliferation of mobile communication devices. However, many existing methods for constructing fingerprint data sets rely on labor-intensive and time-consuming processes of collecting large amounts of data. Additionally, these methods often focus on ideal laboratory environments, rather than considering the practical challenges of large multifloor buildings. To address these issues, we present a novel WiFi domain adversarial graph convolutional network model that can be trained using a small number of labeled site survey data and large amounts of unlabeled crowdsensed WiFi fingerprints. By constructing heterogeneous graphs based on received signal strength indicators (RSSIs) between waypoints and WiFi access points (APs), our model is able to effectively capture the topological structure of the data. We also incorporate graph convolutional networks (GCNs) to extract graph-level embeddings, a feature that has been largely overlooked in previous WiFi indoor localization studies. To deal with the challenges of large amounts of unlabeled data and multiple data domains, we employ a semisupervised domain adversarial training scheme to effectively utilize unlabeled data and align the data distributions across domains. Our system is evaluated using a public indoor localization data set that includes multiple buildings, and the results show that it performs competitively in terms of localization accuracy in large buildings.

As trajectory plays an important role in intelligent transportation systems, achieving trajectory recovery from gradients is of great value. Existing research has shown that federated learning is vulnerable to attacks that recover the original training data from shared gradients. Still, we find that gradients attack is difficult to succeed with trajectory data. Most existing attacks have minimal effectiveness when facing models with higher nonlinearity and temporal-related characteristics, such as destination prediction models. In this article, we propose deep leakage from gradients with mobility prior (DLGMP) algorithm to solve these problems. The proposed DLGMP algorithm leverages the spatiotemporal structural information as the prior mobility knowledge, narrowing down the initial search space sharply and decreasing the difficulty of recovery. We further improve our algorithm with an easily extensible regularization term and an adversarial loss of Wasserstein GAN (WGAN) to help recover accurate and reasonable trajectories. Experiments on three mainstream data sets show good performance of our DLGMP algorithm. We also discuss several possible solutions to prevent gradients attack.

How to generate road conditions from urban GPS trajectory is an important problem in transportation systems. However, this computation process usually suffers from serious missing value problem due to the observation uncertainty or limited reports from crowdsourcing systems. Conventional tensor factorization approaches learn the spatio-temporal dependencies in a collaborative filtering way, which ignores the complex road network structure information and temporal heterogeneity. In this study, we propose a multi-view model with multiple aspects of prior knowledge to impute traffic state computed from a real-world trajectory dataset. More specifically, in the spatial view, rather than focusing on a specific type of road segment, we take the heterogeneity of road network into consideration and model the multiple relations of adjacent road segments. Meanwhile, the temporal pattern is also viewed as a heterogeneous graphical structure that discriminates the weekly/hourly adjacency in the temporal view. Finally, we fuse the above spatio-temporal features to provide a robust estimation under different sparse conditions. Intensive experiments on two types of missing scenarios (i.e., random and non-random) demonstrate that the proposed imputation method outperforms all the other state-of-the-art approaches. In addition, our model represents interpretable patterns for spatio-temporal graph analysis.

Predicting the density and flow of the crowd or traffic at a citywide level becomes possible by using the big data and cutting-edge AI technologies. It has been a very significant research topic with high social impact, which can be widely applied to emergency management, traffic regulation, and urban planning. In particular, by meshing a large urban area to a number of fine-grained mesh-grids, citywide crowd and traffic information in a continuous time period can be represented with 4D tensor (Timestep, Height, Width, Channel). Based on this idea, a series of methods have been proposed to address grid-based prediction for citywide crowd and traffic. In this study, we revisit the density and in-out flow prediction problem and publish a new aggregated human mobility dataset generated from a real-world smartphone application. Comparing with the existing ones, our dataset holds several advantages including large mesh-grid number, fine-grained mesh size, and high user sample. Towards this large-scale crowd dataset, we propose a novel deep learning model called DeepCrowd by designing pyramid architectures and high-dimensional attention mechanism based on Convolutional LSTM. Lastly, thorough and comprehensive performance evaluations are conducted to demonstrate the superiority of the proposed DeepCrowd comparing to multiple state-of-the-art methods.

Recently, forecasting the crowd flows has become an important research topic, and plentiful technologies have achieved good performances. As we all know, the flow at a citywide level is in a mixed state with several basic patterns (e.g., commuting, working, and commercial) caused by the city area functional distributions (e.g., developed commercial areas, educational areas and parks). However, existing technologies have been criticized for their lack of considering the differences in the flow patterns among regions since they want to build only one comprehensive model to learn the mixed flow tensors. Recognizing this limitation, we present a new perspective on flow prediction and propose an explainable framework named ST-ExpertNet, which can adopt every spatial-temporal model and train a set of functional experts devoted to specific flow patterns. Technically, we train a bunch of experts based on the Mixture of Experts (MoE), which guides each expert to specialize in different kinds of flow patterns in sample spaces by using the gating network. We define several criteria, including comprehensiveness, sparsity, and preciseness, to construct the experts for better interpretability and performances. We conduct experiments on a wide range of real-world taxi and bike datasets in Beijing and NYC. The visualizations of the expert's intermediate results demonstrate that our ST-ExpertNet successfully disentangles the city's mixed flow tensors along with the city layout, e.g., the urban ring road structure. Different network architectures, such as ST-ResNet, ConvLSTM, and CNN, have been adopted into our ST-ExpertNet framework for experiments and the results demonstrates the superiority of our framework in both interpretability and performances.

Travel time estimation from GPS trips is of great importance to order duration, ridesharing, taxi dispatching, etc. However, the dense trajectory is not always available due to the limitation of data privacy and acquisition, while the origin-destination (OD) type of data, such as NYC taxi data, NYC bike data, and Capital Bikeshare data, is more accessible. To address this issue, this paper starts to estimate the OD trips travel time combined with the road network. Subsequently, aMulti-taskWeaklySupervisedLearning Framework forTravelTimeEstimation (MWSL-TTE) has been proposed to infer transition probability between roads segments, and the travel time on road segments and intersection simultaneously. Technically, given an OD pair, the transition probability intends to recover the most possible route. And then, the output of travel time is equal to the summation of all segments’ and intersections’ travel time in this route. A novel route recovery function has been proposed to iteratively maximize the current routes’ co-occurrence probability, and minimize the discrepancy between routes’ probability distribution and the inverse distribution of routes’ estimation loss. Moreover, the expected log-likelihood function based on a weakly-supervised framework has been deployed in optimizing the travel time from road segments and intersections concurrently. We conduct experiments on a wide range of real-world taxi datasets in Xi’an and Chengdu and demonstrate our method's effectiveness on route recovery and travel time estimation.

The outbreak of coronavirus disease (COVID-19) has swept across more than 180 countries and territories since late January 2020. As a worldwide emergency response, governments have implemented various measures and policies, such as self-quarantine, travel restrictions, work from home, and regional lockdown, to control the spread of the epidemic. These countermeasures seek to restrict human mobility because COVID-19 is a highly contagious disease that is spread by human-to-human transmission. Medical experts and policymakers have expressed the urgency to effectively evaluate the outcome of human restriction policies with the aid of big data and information technology. Thus, based on big human mobility data and city POI data, an interactive visual analytics system called Epidemic Mobility (EpiMob) was designed in this study. The system interactively simulates the changes in human mobility and infection status in response to the implementation of a certain restriction policy or a combination of policies (e.g., regional lockdown, telecommuting, screening). Users can conveniently designate the spatial and temporal ranges for different mobility restriction policies. Then, the results reflecting the infection situation under different policies are dynamically displayed and can be flexibly compared and analyzed in depth. Multiple case studies consisting of interviews with domain experts were conducted in the largest metropolitan area of Japan (i.e., Greater Tokyo Area) to demonstrate that the system can provide insight into the effects of different human mobility restriction policies for epidemic control, through measurements and comparisons.

Citywide crowd prediction can be of great importance for emergency management, traffic regulation, and urban planning. By meshing a large urban area to a number of fine-grained mesh-grids as illustrated in Fig. 1 , citywide crowd in a continuous time period can be represented with a four-dimensional tensor ${\mathbb{R}^{Timestep{\text{ }}p \times {\text{ }}Height{\text{ }} \times {\text{ }}Width{\text{ }} \times {\text{ }}Channel{\text{ }}}}$ in an analogous manner to video data, where each Timestep can be seen as one video frame, Height , Width is two-dimensional index for mesh-grids, and each Channel stores an aggregated scalar value for each mesh-grid. Specifically, given historical observations of crowd density and in-out flow x d = d 1 ,…,d t , xf = f 1 ,…, f t at timestamp t , we aim to build prediction models for the next-step density and in-out flow y d = d t +1, yf = f t +1, where y d means how many people will be in each mesh-grid at the next timestamp, and yf means how many people will flow into or out from each mesh-grid in next time interval. Al-though many deep models [1] – [6] have been proposed to address such tasks, their actual effects are still not well validated on large-scale and high-quality datasets. The datasets used in most of the works so far are originally generated based on taxi or bicycle trip data, which don’t cover and reflect the citywide crowd density and flow. Thus, we first publish new crowd flow data called BousaiTYO and BousaiOSA [7] . These new datasets are created using the GPS log data collected from a popular smartphone app of Yahoo! Japan Corporation, which can well reflect the real-world crowd flow in Tokyo and Osaka. As shown by Table 1 , our dataset has: (1) larger spatial area; (2) finer mesh size; (3) higher user sample.

Friend recommendation from human mobility is a vital real-world application of location-based social networks (LBSN). It is necessary to recognize patterns from human mobility to assist friend recommendation because previous works have shown complex relations between them. However, most of previous works either modelled social networks and user trajectories separately, or only used classical simple graph-based methods with an edge linking two nodes that cannot fully model the complex data structure of LBSN. Inspired by the fact that hyperedges can connect multiple nodes of different types, we model user trajectories and check-in records as hyperedges in a novel heterogeneous LBSN hypergraph to represent complex spatio-temporal information. And then, we design a type-specific attention mechanism for an end-to-end trainable heterogeneous hypergraph neural network (HHGNN) with supervised contrastive learning, which can learn hypergraph node embedding for the next friend recommendation task. At last, our model HHGNN outperforms the state-of-the-art methods on four real-world city datasets, while ablation studies also confirm the effectiveness of each model part.

Finding out similar regions between cities is important to a variety of real-world applications, such as point-of-interest recommendations, site selection, and travel guidance. With the help of the increasing number of location-based social network users, we can measure the intercity similarity from a new perspective of spatiotemporal characteristics of human mobility. In this paper, we developed an interactive intercity regional similarity explorer (IRSE) that 1) visualizes regional spatiotemporal human mobility features, 2) searches similar region candidates in the target city, and 3) explores the regional similarity from different views in a quantitative and illustrative way. In this paper, we show how our system can be useful in exploring regional similarity across cities in the world by use cases, which will interest various users from different countries in the demonstration session. Demo available at: https://bit.ly/3OjwnGu

Knowing "what is happening" and "what will happen" of the mobility in a city is the building block of a data-driven smart city system. In recent years, mobility digital twin that makes a virtual replication of human mobility and predicting or simulating the fine-grained movements of the subjects in a virtual space at a metropolitan scale in near real-time has shown its great potential in modern urban intelligent systems. However, few studies have provided practical solutions. The main difficulties are four-folds: 1) the daily variation of human mobility is hard to model and predict; 2) the transportation network enforces a complex constraints on human mobility; 3) generating a rational fine-grained human trajectory is challenging for existing machine learning models; and 4) making a fine-grained prediction incurs high computational costs, which is challenging for an online system. Bearing these difficulties in mind, in this paper we propose a two-stage human mobility predictor that stratifies the coarse and fine-grained level predictions. In the first stage, to encode the daily variation of human mobility at a metropolitan level, we automatically extract citywide mobility trends as crowd contexts and predict long-term and long-distance movements at a coarse level. In the second stage, the coarse predictions are resolved to a fine-grained level via a probabilistic trajectory retrieval method, which offloads most of the heavy computations to the offline phase. We tested our method using a real-world mobile phone GPS dataset in the Kanto area in Japan, and achieved good prediction accuracy and a time efficiency of about 2 min in predicting future 1h movements of about 220K mobile phone users on a single machine to support more higher-level analysis of mobility prediction.

Individual trajectory prediction is a sequential forecasting task, which uses a moving agent's past trajectory to predict possible future trajectories. Existing work trains one predictor for all users, while few studies consider a personalized predictor that automatically extracts the personal trajectory characteristics for each individual. Also, individual trajectories are highly random and in-homogeneous, resulting in some real target locations are not in the training data set totally, making the model difficult to converge. To address above difficulties, we propose a pre-trained trajectory prediction model via meta-learning, which not only can learn a more generalized initialization parameters to extract the trajectory features of multiple individuals, but also solve the problem of in-homogeneous distribution using pre-trained grid-based classification.

Predicting the density and flow of the crowd at a citywide level is significant for city management. By meshing a large urban area to a number of fine-grained mesh-grids, citywide crowd and traffic information in a continuous time period can be represented with 4D tensor (Timestep, Height, Width, Channel). Based on this, we revisit the density and in-out flow prediction problem and publish a new aggregated human mobility dataset generated from a real-world smartphone application. Compared with the existing ones, our dataset has larger mesh-grid number, finer-grained mesh size, and higher user sample. Towards such kind of large-scale crowd dataset, we propose a novel deep learning model called DeepCrowd by designing pyramid architectures and high-dimensional attention mechanism based on Convolutional LSTM. Both the datasets and codes are made available at https://github.com/deepkashiwa20/DeepCrowd.

This paper introduces a cross-scale attention based end-to-end learning framework for tree crown detection via UAV imagery. Given that UAV images covered a large forests, the illumination variations, shadow obstacles and texture repetition always lead to inaccurate tree crown detection results. We introduce a cross-scale attention based mechanism to address the above issues, enabling the tree crown detection framework to reason about the RGB texture information and depth information introduced by the automatically generated depth map jointly. Compared to traditional image based tree crown detection methods, our approach learns prior over geometrical structure information from the real 3D world, which is robust to the texture repetition and small tree crowns. The experimental results demonstrated that the proposed approach outperforms the traditional CNN based method.

Estimating the travel time of a given path is an important topic for the intelligent transportation system and serves as the foundation for various real-world applications. However, building an estimation model for such a data-driven task requires a large amount of mobile users' trajectory data which directly relates to their privacy and thus is less likely to be shared. Therefore, we propose GPF-TTE, Generative Personalized Federated Learning Framework for Travel Time Estimation (poster version of our previous work [1]) based on the issue of privacy protection for the mobile user group, in which 1) utilizes the federated learning approach, allowing private data to be kept on client devices while training, 2) apart from sharing a base model, we also adapt a fine-tuned personalized model for each client to study their personal driving habits, making up for the residual error caused by the prediction of the localized global model (the base model in local device), and 3) the cloud server aggregates localized models into the global model as a generative model to infer the future road traffic state.

Estimating the travel time of a path is an essential topic for the intelligent transportation system. It serves as the foundation for real-world applications, such as traffic monitoring, route planning, and taxi dispatching. However, building a model for such a data-driven task requires a large amount of users’ travel information, which closely relates to their privacy and, thus, is less likely to be shared. The not independent and identically distributed (Non-IID) trajectory data across data owners also make a predictive model extremely challenging to be personalized if we directly apply federated learning. Finally, previous work on travel time estimation (TTE) does not consider the real-time traffic state of roads, which we argue, can significantly influence the prediction. To address the above challenges, we introduce GOF-TTE for the mobile user group, generative online federated learning framework for TTE, which 1) utilizes the federated learning approach, allowing private data to be kept on client devices while training, and designs the global model as an online generative model shared by all clients to infer the real-time road traffic state and 2) apart from sharing a base model at the server, adapts a fine-tuned personalized model for every client to study their personal driving habits, making up for the residual error made by localized global model prediction. We also employ a simple privacy attack to our framework and implement the differential privacy mechanism to guarantee privacy safety further. Finally, we conduct experiments on two real-world public taxi data sets of DiDi Chengdu and Xi’an. The experimental results demonstrate the effectiveness of our proposed framework.

Event crowd management has been a significant research topic with high social impact. When some big events happen such as an earthquake, typhoon, and national festival, crowd management becomes the first priority for governments (e.g., police) and public service operators (e.g., subway/bus operator) to protect people’s safety or maintain the operation of public infrastructures. However, under such event situations, human behavior will become very different from daily routines, which makes prediction of crowd dynamics at big events become highly challenging, especially at a citywide level. Therefore in this study, we aim to extract the “deep” trend only from the current momentary observations and generate an accurate prediction for the trend in the short future, which is considered to be an effective way to deal with the event situations. Motivated by these, we build an online system called DeepUrbanEvent, which can iteratively take citywide crowd dynamics from the current one hour as input and report the prediction results for the next one hour as output. A novel deep learning architecture built with recurrent neural networks is designed to effectively model these highly complex sequential data in an analogous manner to video prediction tasks. Experimental results demonstrate the superior performance of our proposed methodology to the existing approaches. Lastly, we apply our prototype system to multiple big real-world events and show that it is highly deployable as an online crowd management system.

Forecasting incident occurrences (e.g. crime, EMS, traffic accident) is a crucial task for emergency service providers and transportation agencies in performing response time optimization and dynamic fleet management. However, such events are by nature rare and sparse, which causes the label imbalance problem and inferior performance of models relying on data sufficiency. The existing studies circumvent, instead of truly solving, this issue by defining the incident prediction problem in a coarse-grained temporal (e.g. daily) setting, which leaves the proposed models unrobust to fine-grained dynamics and trivial for the real-world decision making. In this paper, we tackle the temporally fine-grained incident prediction problem in a sparse setting by explicitly exploiting the behind-the-scene chainlike triggering mechanism. Moreover, this chain effect roots in multiple domains (i.e. spatial, categorical), which further entangles with the temporal dimension and happens to be time-variant. To be specific, we propose a novel deep learning framework, namely Spatio-Temporal-Categorical Graph Neural Networks (STC-GNN), to handle the multidimensional and dynamic chain effect for performing fine-grained multi-incident co-prediction. Extensive experiments on three real-world city-level incident datasets verify the insightfulness of our perspective and effectiveness of the proposed model.

Transportation demand forecasting is a topic of large practical value. However, the model that fits the demand of one transportation by only considering the historical data of its own could be vulnerable since random fluctuations could easily impact the modeling. On the other hand, common factors like time and region attribute, drive the evolution demand of different transportation, leading to a co-evolving intrinsic property between different kinds of transportation. In this work, we focus on exploring the co-evolution between different modes of transport, e.g., taxi demand and shared-bike demand. Two significant challenges impede the discovery of the co-evolving pattern: (1) diversity of the co-evolving correlation, which varies from region to region and time to time. (2) Multi-modal data fusion. Taxi demand and shared-bike demand are time-series data, which have different representations with the external factors. Moreover, the distribution of taxi demand and bike demand are not identical. To overcome these challenges, we propose a novel method, known as co-evolving spatial temporal neural network (CEST). CEST learns a multi-view demand representation for each mode of transport, extracts the co-evolving pattern, then predicts the demand for the target transportation based on multi-scale representation, which includes fine-scale demand information and coarse-scale pattern information. We conduct extensive experiments to validate the superiority of our model over the state-of-art models.

Accurate human path forecasting in complex and crowded scenarios is critical for collision avoidance of autonomous driving and social robots navigation. It still remains as a challenging problem because of dynamic human interaction and intrinsic multimodality of human motion. Given the observation, there is a rich set of plausible ways for an agent to walk through the circumstance. To address those issues, we propose a spatio-temporal model that can aggregate the information from socially interacting agents and capture the multimodality of the motion patterns. We use mixture density functions to describe the human path and predict the distribution of future paths with explicit density. To integrate more factors to model interacting people, we further introduce a coordinate transformation to represent the relative motion between people. Extensive experiments over several trajectory prediction benchmarks demonstrate that our method is able to forecast various plausible futures in complex scenarios and achieves state-of-the-art performance.

Nowadays, GPS devices have increased explosively and produced huge amounts of trajectory data related to people's outgoing. Through those big location data, many researches aim to analyze human mobility for urban development, such as human movement prediction/modeling, POI (Point-Of-Interest) recommendation. However, trajectory data only contains timestamp and location information. The intention of human movement is not explicit so that it is hard to understand why people go to somewhere. The intention prior to the activity could be of great significance for analyzing and predicting human mobility, which has not been taken into consideration by the existing researches until the present. Thus, in this study, we propose a brand-new concept called human intentional mobility, aiming to employ intention information to predict people's outgoing. We carefully utilize user's search query to sense his intention as well as the intensity. For instance, if a user searches a certain POI for many times in a short period, it will represent a relatively high intention to go there. Then, to fully utilize this intention representation for predicting whether user will visit searched POI or not, we specially design Dual Sequential Interaction Network (DualSIN) as a novel and unique deep-learning model, which can effectively capture the sophisticated interactions among two kinds of sequential information (i.e., search sequence and mobility sequence) and typical categorical information (i.e., user attributes). Last, we evaluate our model on real-world dataset collected from Yahoo! Japan portal application, and demonstrate that it can achieve superior satisfactory performances to the-state-of-the-art models on multiple POI search queries.

Extracting identifiable information from human trajectories is a fundamental task in many location-based services (LBS). However, various mobility patterns underlain in human trajectories are difficult to model by existing models. Moreover, we could hardly define a clear user set for user identification because the set of users are dynamic and changing everyday. Bearing these in mind, we apply a deep multiple instance learning method to handle the multimodal mobility patterns in a weak-supervised learning way, and address the dynamic user set problems via a pairwise loss with negative sampling. We utilize a multi-head attention mechanism to automatically extract multiple aspects and match the corresponding information between query trajectories and historical trajectories. Our method shows a good identification accuracy on three human GPS trajectory data sets comparing with baseline methods.

Big human mobility data are being continuously generated through a variety of sources, some of which can be treated and used as streaming data for understanding and predicting urban dynamics. With such streaming mobility data, the online prediction of short-term human mobility at the city level can be of great significance for transportation scheduling, urban regulation, and emergency management. In particular, when big rare events or disasters happen, such as large earthquakes or severe traffic accidents, people change their behaviors from their routine activities. This means people's movements will almost be uncorrelated with their past movements. Therefore, in this study, we build an online system called DeepUrbanMomentum to conduct the next short-term mobility predictions by using (the limited steps of) currently observed human mobility data. A deep-learning architecture built with recurrent neural networks is designed to effectively model these highly complex sequential data for a huge urban area. Experimental results demonstrate the superior performance of our proposed model as compared to the existing approaches. Lastly, we apply our system to a real emergency scenario and demonstrate that our system is applicable in the real world.

Globalization leads to more and more outbound behavior. Understanding these behavior is of great significance to both outbound and inbound countries in a lot of fields. However, the previous research on outbound behavior mainly unitizes questionnaires or purely analyzes the statistical data, these data are either too difficult to acquire or unable to directly analyze public opinion. This paper aim to analyze outbound behavior through large volume of text and location information published on social network services (SNS) and look into a case study of Chinese people in Japan with the data acquired from Chinese microblog website, Sina Weibo. For text analysis, we utilize an adapted latent Dirichlet allocation model to extract the topics in text and discuss the spatial and temporal distribution of the topics. The experiment proves that the topics represent outbound behavior of tourists and foreign visitors can be found and explained in the experiment dataset, which indicates that SNS data is valid for analyzing outbound behavior.

There are two broad categories of citywide human mobility, routine, composed of daily or periodic travel, and rare, which occurs during events such as the Olympic Games or natural disasters. State-of-the-art studies have shown that routine mobility patterns can be modeled stochastically, while rare human mobility modeling, essential to a variety of urban computing scenarios, such as emergency management and traffic regulation, is a much more challenging and understudied problem. Instead of training a rare-event-specific human mobility model, which suffers from the particularity of the rare events, in this paper we provide a new insight into rare events and propose a novel algorithm, CityCoupling, which establishes an intercity spatial mapping that uses human mobility in one city as input and reproduces human mobility in another city. More intuitively, we attempt to answer the question "What if this rare event happened in another city?". To find the optimal intercity spatial mapping, we utilize an expectation-maximization algorithm to estimate a probabilistic geographical correspondence matrix by regarding intercity trajectory matching as latent variables. Thereafter, a Gibbs sampling-based multiple hidden Markov model generates simulated trajectories. We apply our approach to a large mobile phone GPS dataset in Japan and determine the spatial mapping between Tokyo and Osaka to transfer the human mobility at the Great Eastern Japan Earthquake in Tokyo, which was heavily affected, to simulate what might have occurred if Osaka had been struck by the earthquake. We conduct the evaluation by assuming that New Year's Countdown is a rare event that occurs simultaneously in both Tokyo and Osaka, and thus we quantitatively compare our simulation with the ground truth in Osaka.

Human movements are difficult to predict, especially, when we consider rare behaviors that deviate from normal daily routines. By tracing the behavior of a person over a long period, we can model their daily routines and predict periodical behaviors, whereas rare behaviors, such as participating in the New Year's Eve countdown, can hardly be predicted readily and thus they have usually been treated as outliers of the daily routines in most existing studies. However, for scenarios such as emergency management or intelligent traffic regulation, we are more interested in rare behaviors than daily routines. Using human mobility Big Data, the rare behavior of each individual in a social crowd is no longer rare and thus it may be predicted when we analyze the crowd behavior at a citywide level. Therefore in this study, instead of predicting movement based on daily routines, we make short-term predictions based on the recent movement observations. We propose a novel model called CityMomentum as a predicting-by-clustering framework for sampling future movement using a mixture of multiple random Markov chains, each of which is a Naive Movement Predictive model trained with the movements of the subjects that belong to each cluster. We apply our approach to a big mobile phone GPS log dataset and predict the short-term future movements, especially during the Comiket 80 and New Year's Eve celebration. We evaluate our prediction by a Earth Mover Distance (EMD) based metric, and show our approach accurately predicts the crowd behavior during the rare crowd events, which makes an early crowd event warning and regulation possible in the emergent situations.

People flow at a citywide level is in a mixed state with several basic patterns (e.g. commuting, working, commercial), and it is therefore difficult to extract useful information from such a mixture of patterns directly. In this paper, we proposed a novel tensor factorization approach to modeling city dynamics in a basic life pattern space (CitySpectral Space). To obtain the CitySpectrum, we utilized Non-negative Tensor Factorization (NTF) to decompose a people flow tensor into basic life pattern tensors, described by three bases i.e. the intensity variation among different regions, the time-of-day and the sample days. We apply our approach to a big mobile phone GPS log dataset (containing 1.6 million users) to model the fluctuation in people flow before and after the Great East Japan Earthquake from a CitySpectral perspective. In addition, our framework is extensible to a variety of auxiliary spatial-temporal data. We parametrize a people flow with a spatial distribution of the Points of Interest (POIs) to quantitatively analyze the relationship between human mobility and POI distribution. Based on the parametric people flow, we propose a spectral approach for a site-selection recommendation and people flow simulation in another similar area using POI distribution.