Multivariate time series (MTS) anomaly detection identifies abnormal patterns where each timestamp contains multiple variables. Existing MTS anomaly detection methods fall into three categories: reconstruction-based, prediction-based, and classifier-based methods. However, these methods face three key challenges: (1) Unsupervised learning methods, such as reconstruction-based and prediction-based methods, rely on error thresholds, which can lead to inaccuracies; (2) Semi-supervised methods mainly model normal dataand often underuse anomaly labels, limiting detection of subtle anomalies; (3) Supervised learning methods, such as classifier-based approaches, often fail to capture local relationships, incur high computational costs, and are constrained by the scarcity of labeled data. To address these limitations, we proposeMoon, a supervised modality conversion-based multivariate time series anomaly detection framework.Moonenhances the efficiency and accuracy of anomaly detection while providing detailed anomaly analysis reports. First,Moonintroduces a novel multivariate Markov Transition Field (MV-MTF) technique to convert numeric time series data into image representations, capturing relationships across variables and timestamps. Since numeric data retains unique patterns that cannot be fully captured by image conversion alone,Moonemploys a Multimodal-CNN to integrate numeric and image data through a feature fusion model with parameter sharing, enhancing training efficiency. Finally, a SHAP-based anomaly explainer identifies key variables contributing to anomalies, improving interpretability. Extensive experiments on six real-world MTS datasets demonstrate thatMoonoutperforms six state-of-the-art methods by up to 93% in efficiency, 4% in accuracy and, 10.8% in interpretation performance.

Trajectory data analysis benefits numerous real-world applications and has attracted substantial attention from the research community. With the rapid proliferation of IoT devices and the emergence of edge computing, there has been an increasing demand for efficient trajectory data analytics in edge environments. However, most existing trajectory analysis systems are designed for cloud-based architectures, which face significant limitations in edge settings. These include resource constraints, dynamic network conditions, and inefficient query handling, leading to sub-optimal performance in edge scenarios. To fill this gap, we propose TrajEdge, an efficient and lightweight framework for trajectory data analysis in edge environments. Implementing TrajEdge requires overcoming obstacles posed by limited resources and the dynamic nature of edge networks. To achieve this, we design a novel trajectory composite compression algorithm that delivers high compression ratios, significantly reducing storage pressure on edge devices. Additionally, we introduce three coflow control strategies optimized for varying network conditions, enabling higher system throughput. To further enhance the efficiency of trajectory queries, we develop a spatiotemporal-aware trie-based peer-to-peer (P2P) index. Experimental evaluations on two real-world datasets and one larger synthetic dataset demonstrate that TrajEdge achieves remarkable performance improvements: more than 200 × gains in storage and query efficiency, up to 64% increases in network throughput, compression ratios of up to 95%, and exceptional scalability compared to the state-of-the-art systems. Our source code is available at https://github.com/ZJU-DAILY/TrajEdge.

Spatio-temporal data prediction is a fundamental task in urban computing, benefiting a variety of real-life applications such as traffic forecasting and environmental monitoring. Due to the dynamic and time-involving nature of spatio-temporal data, researchers have increasingly emphasized online prediction. However, existing approaches (e.g., URCL) typically rely on data-replay strategies, which require storing large volumes of historical data to frequently update their models with new inputs. These methods impose substantial costs, including frequent buffer construction, high storage requirements, and increased training complexity. Furthermore, the single-pass nature of online data, combined with the constrained resources of online environments, highlights the urgent need for more efficient and lightweight solutions for online spatio-temporal prediction. To address these challenges, we propose Storm, a knowledge distillation driven continual learning framework. Storm introduces Dynamic Knowledge Distillation (DKD), leveraging an ever-evolving teacher model to train an effective student model. To optimize efficiency, Storm employs a Mixture-of-Experts (MoE) mechanism, which dynamically switches between the original training mode and the DKD mode. This hybrid design enables low-cost online learning while addressing the stabilityplasticity dilemma. To fully leverage single-pass online data, Storm integrates effective data augmentation methods tailored to the dynamic nature of spatio-temporal data. Moreover, Storm incorporates a Gradual Parameter Freezing (GPF) module to progressively reduce computational costs during online training. Extensive experiments conducted on four real-world datasets, evaluated across short-term, medium-term, and long-term prediction horizons, demonstrate the superiority of Storm. Specifically, Storm: (i) provides a general online training extension for various offline spatio-temporal models, and (ii) achieves remarkable improvements, e.g., up to 14.24% accuracy gains while requiring only 0.3% of the training and inference time compared to the state-of-the-art URCL framework. The source code is publicly available at https://github.com/ZJU-DAILY/Storm.

Spatio-temporal traffic series prediction is essential in intelligent transportation systems, benefiting various applications such as route planning, vehicle dispatching, and congestion prediction. To tackle privacy leakage aroused by centralized forecasting methods, Federated Learning (FL), a privacy-preserving approach for decentralized model training into disjointed federated clients, has garnered widespread adoption in numerous traffic prediction endeavors. However, existing FL-based approaches ignore spatio-temporal heterogeneity among federated clients, including spatial feature skew, temporal coverage skew, and data quality skew. This makes them inapplicable and unsuitable to real-world scenarios and exhibits subpar prediction performance. To this end, we perform the first study of heterogeneous-aware traffic prediction in the federated environment, proposing a unified and effective framework named Fed4TP. It offers general federated capability for various centralized forecast models, supporting flow, speed, and occupancy prediction tasks. To address spatial feature heterogeneity, Fed4TP develops multi-dimensional personalized federated learning with positive samples contrastive learning for clustering to achieve personalized aggregation and global sharing across diverse clients. To overcome temporal coverage heterogeneity, Fed4TP designs a time window-based federated training mechanism, sequentially training client models and learning missed traffic information with varying time coverage. To tackle data quality heterogeneity, Fed4TP introduces a dual-driven method, i.e., global detection and local denoising, to improve client data quality. Extensive experiments on 4 real-life datasets verify the effectiveness and scalability superiority of Fed4TP in various federated-based traffic prediction tasks, compared with 24 well-known and state-of-the-art baselines. The source code and data of this work are available at https://github.com/ZJU-DAILY/Fed4TP.

With the growing availability of large data repositories within and across organizations, it is becoming feasible to selectively acquire data in the wild for data augmentation, tailored to specific downstream tasks. However, current methodologies concentrate primarily on single-dimensional augmentation tasks for tabular data, such as increasing the number of data points or enriching features. These approaches are not designed for trajectory data, which cannot fully utilize its spatio-temporal characteristics, resulting in suboptimal performance.

Trajectory anomaly detection aims to identify patterns in trajectory data that deviate significantly from normal behavior, such as taxi detours, and plays a crucial role in urban computing. However, real-world trajectories are inherently complex, containing diverse anomalies and unavoidable noise. Existing research mainly focuses on coarse-grained trajectory anomalies, such as detour and switch anomalies, while paying limited attention to fine-grained trajectory anomalies, such as time and loop anomalies. Furthermore, they tend to disregard the impact of inherent noise in trajectories. As a result, there remains a gap in developing robust models with strong generalization capabilities to effectively detect fine-grained trajectory anomalies, even in noisy environments.

Abstract In daily life, there is an increasing demand for efficient management and analysis of spatio-temporal data. However, current systems struggle to balance multi-functionality, scalability, and computational efficiency in this domain. To address this challenge, we introduce PostMan: a productive spatio-temporal data management system. PostMan is based on Apache Spark and Apache Hadoop HDFS. It extensively, efficiently, and scalably supports spatio-temporal data types and operators across multiple API levels. To realize effective data management and analysis, PostMan designs the unified partition management and hybrid index. Based on this, PostMan has designed and implemented a variety of optimization strategies for vector and raster operators. PostMan also introduces a two-phase static partitioning (TPSP) method to maintain load balance before and after partition filtering during the query process. In the first phase, partitions are generated using an enhanced R*-Tree algorithm, while the second phase allocates partitions by modeling the task as an optimization problem solved through greedy algorithms. For faster computation, PostMan introduces processes and program interfaces for GPU accelerated spatio-temporal operators in Spark. Moreover, extensive evaluations using real-world datasets show PostMan’s notable efficiency and scalability advantages (e.g., 13%-36% improvement) over baseline systems, as well as their constituent techniques. Finally, PostMan has been deployed on the public cloud in a Software as a Service (SaaS) model, garnering substantial attention from customers.

Vector search is widely employed in recommendation systems, search engines, etc. With the explosive growth of online data and streaming processing engines, streaming vector search has attracted increasing research attention. However, prevailing vector search systems like Vearch, Vespa, and Milvus typically operate as external batch services for streaming processing requirements, resulting in sub-optimal performance for streaming processing scenarios. In this paper, we propose VStream, a distributed streaming vector search system. Implementing such a system is non-trivial, raising three technical challenges in streaming adaptability, system scalability, and real-time response. Specifically, VStream offers a dynamic partitioner that adapts to data distribution changes in vector streams. Additionally, VStream features an effective hierarchical storage architecture facilitated by streaming state management, enabling a hybrid of four-level storage media with diverse access speeds and targets. Furthermore, VStream utilizes dynamic hot-cold patterns, such as access frequency, in the streaming vector data, incorporating a specialized hot-cold separation mechanism to enhance query efficiency. Extensive experiments prove that VStream outperforms existing vector search systems, e.g., achieving 251–373× improvements in query efficiency, 2.2–2.5× savings in CPU usage, and 1.5–2.0× reductions in memory overhead.

Trajectory similarity computation in road networks is crucial for data analytics. However, both non-learning-based and learning-based methods face challenges. First, they suffer from low accuracy due to manual parameter selection for model training and the omission of key spatio-temporal features in road networks. Second, they have low efficiency, stemming from the high time complexity of similarity computation and the time-consuming training process. Third, learning-based methods struggle with poor model generality due to the small size of available training samples. To address these challenges, we propose an effective and efficient trajectory similarity learning framework for road networks, called SimRN. To our knowledge, SimRN is the first deep reinforcement learning (DRL) approach for trajectory similarity computation. Specifically, SimRN consists of three key modules: the spatio-temporal prompt information extraction (STP) module, the trajectory representation based on DRL (TrajRL) module, and the graph contrastive learning (GCL) module. The STP module captures spatio-temporal features from road networks to improve the training of the trajectory representation. The TrajRL module automatically selects optimal parameters and enables parallel training, improving both trajectory representation and the efficiency of similarity computations. The GCL module employs a self-supervised contrastive learning paradigm to generate sufficient samples while preserving spatial constraints and temporal dependencies of trajectories. Extensive experiments on two real-world datasets, compared with three state-of-the-art methods, show that SimRN: (i) improves accuracy by 20%–40%, (ii) achieves speedups of 2–4x, and (iii) demonstrates strong generality, enabling effective similarity learning with very small sample sizes.

Long-term time series forecasting has gained significant attention in recent years due to its widely-application in various fields. Transformer-based models have gained popularity for the ability to capture long-sequence interactions. However, these models are limited in real-world use because of the memory consumption and computation explosion. The CNN-based models are also one of the main models used for time series prediction, but their performance has always been inferior to the transformer-based models in previous works. We have reconsidered the role of CNN components and redefined the way CNN basic components are used for time series prediction. In addition, the time lags information between periods in the time series is important. Unfortunately, existing works lack consideration of this classic but important information. Motivated by these factors, we propose a fast yet effective CNN model with time lags for multivariate long-term time series forecasting, named LagCNN. Specifically, the time series is transformed into lag-patches to capture the correlation between periods. Then, a fast CNN model is performed in the feature dimension rather than the time dimension like most previous works do. Meanwhile, information aggregation is performed in the time dimension to extract complex temporal patterns. LagCNN significantly outperforms state-of-the-art on multiple publicly available datasets. One step further, LagCNN exhibits significant efficiency advantages over the most efficient Transformer model (PatchTST), resulting in a significant reduction in memory usage (4.4×) and runtime (10.7×).

Trajectory streams are being generated from location-aware devices, such as smartphones and in-vehicle navigation systems. Due to the sensitive nature of the location data, directly sharing user trajectories suffers from privacy leakage issues. Local differential privacy (LDP), which perturbs sensitive data on the user side before it is shared or analyzed, emerges as a promising solution for private trajectory stream collection and analysis. Unfortunately, existing stream release approaches often neglect the rich spatial-temporal context information within trajectory streams, resulting in suboptimal utility and limited types of downstream applications. To this end, we propose RetraSyn, a novel real-time trajectory synthesis framework, which is able to perform on-the-f1y trajectory synthesis based on the mobility patterns privately extracted from users' trajectory streams. Thus, the downstream trajectory analysis can be performed on the high-utility synthesized data with privacy protection. We also take the genuine behaviors of real-world mobile travelers into consideration, ensuring authenticity and practicality. The key components of RetraSyn include the global mobility model, dynamic mobility update mechanism, real-time synthesis, and adaptive allocation strategy. We conduct extensive experiments on multiple real-world and synthetic trajectory datasets under various location-based utility metrics, encompassing both streaming and historical scenarios. The empirical results demonstrate the superiority and versatility of our proposed framework.

Contact tracing query (CTQ) plays a crucial role in the prevention of epidemic diseases. In real-world applications, user trajectory, encompassing a wealth of sensitive information, is typically dispersed across various devices or organizations. Consequently, safeguarding user privacy becomes imperative in the context of CTQ. Simultaneously, for effective epidemic control, it is essential to identify contacts efficiently and accurately, enabling prompt implementation of necessary measures. However, existing CTQ studies face limitations as they struggle to concurrently meet the demands of privacy, accuracy and efficiency. This constraint impedes their practical application in real-world scenarios. To this end, we define the Federated Contact Tracing Query (F-CTQ) problem and propose the FedCTQ framework based on hierarchical federation. To the best of our knowledge, this is the first solution grounded in federation, offering a simultaneous fulfillment of privacy, accuracy and efficiency requirements. Specifically, to ensure the privacy of F-CTQ, we introduce a meticulously designed binary-based secret-sharing (BSS) scheme, which delivers an effective privacy guarantee for user data while preserving the accuracy of the query results. Concurrently, to enhance the efficiency of F-CTQ, we propose a binary-based distance tree (DistTree) index that maximizes computational resources for parallel queries. Based on DistTree, FedCTQ enables nearly the real-time and accurate execution of F-CTQ. Extensive experiments on four datasets demonstrate the superiority of FedCTQ, showcasing a remarkable performance improvement ranging from$4.7\times$to$14.8\times$over state-of-the-art approaches.

Accurate vehicle trajectory recovery enables providing indispensable data foundations in intelligent urban transportation. However, existing methods face two challenges: i) the inability to process city-wide vehicle trajectories, and ii) the dependence on a substantial amount of accurate GPS trajectories for model training, leading to poor generalization ability. To address these issues, we propose a novel trajectory recovery system based on vehicle snapshots captured by traffic cameras, named TrajRecovery. TrajRecovery consists of three main components: i) Preprocessor processes traffic cameras and vehicle snapshots to provide necessary data for trajectory recovery; ii) Spatial Transfer Probabilistic Model (STPM) integrates road conditions and driver behavior to compute turning probability at intersections; iii) Trajectory Generator utilizes the output probabilities from STPM to recover a continuous and most likely complete trajectory. We evaluate TrajRecovery on two real datasets from a city in China, demonstrating substantial performance gains compared to state-of-the-art methods. Furthermore, our system is deployed in practical applications at Huawei Company, achieving extraordinary profits in business scenarios.

Time series compression encodes the information in a time-ordered sequence of data points into fewer bits, thereby reducing storage costs and possibly other costs. Compression methods are either general or XOR-based. General compression methods are time-consuming and are not suitable in streaming scenarios, while XOR-based methods are unable to consistently maintain high compression ratios. Further, existing methods compress the integer and decimal parts of floating-point values as a whole, thus disregarding the different characteristics of the two parts. We propose Camel , a new compression method for floating-point time series with the goal of advancing the compression ratios and efficiency achievable. Camel compresses the integer and decimal parts of the double-precision floating-point numbers in time series separately; and instead of performing XOR operations on values using their previous value, Camel identifies values that enable higher compression ratios. Camel also includes means of indexing compressed data, thereby making it possible to query compressed data efficiently. We report on an empirical study of Camel and 11 lossless and 6 lossy compression methods on 22 public datasets and three industrial datasets from AliCloud. The study offers evidence that Camel is capable of outperforming existing methods in terms of both compression ratio and efficiency and is capable of excellent compression performance on both time series and non-time series data.

With the widespread use of GPS-enabled devices and services, trajectory data fuels services in a variety of fields, such as transportation and smart cities. However, trajectory data often contains errors stemming from inaccurate GPS measurements, low sampling rates, and transmission interruptions, yielding low-quality trajectory data with negative effects on downstream services. Therefore, a crucial yet tedious endeavor is to assess the quality of trajectory data, serving as a guide for subsequent data cleaning and analyses. Despite some studies addressing general-purpose data quality assessment, no studies exist that are tailored specifically for trajectory data. To more effectively diagnose the quality of trajectory data, we propose T-Assess, an automated trajectory data quality assessment system. T-Assess is built on three fundamental principles: i) extensive coverage, ii) versatility, and iii) efficiency. To achieve comprehensive coverage, we propose assessment criteria spanning validity, completeness, consistency, and fairness. To provide high versatility, T-Assess supports both offline and online evaluations for full-batch trajectory datasets as well as real-time trajectory streams. In addition, we incorporate an evaluation optimization strategy to achieve assessment efficiency. Extensive experiments on four real-life benchmark datasets offer insight into the effectiveness of T-Assess at quantifying trajectory data quality beyond the capabilities of state-of-the-art data quality systems.

With the explosive ubiquity of GPS-equipped devices, e.g., mobile phones, vehicles, and vessels, a massive amount of real-time, unbounded, and varying-sampling trajectory streams are being generated continuously. Clustering trajectory streams is useful in real-life applications, such as traffic congestion prediction, crowd flow detection, and moving behavior study. Although several sliding-window based algorithms (that adopt the classic two-phases online-offline processing framework) are proposed for trajectory stream clustering, three challenges exist to meet ever-increasing application demands for effective, efficient, and scalable online clustering: i) How to effectively model unbounded trajectory streams in the online settings for effective clustering? ii) How to achieve truly real-time online processing? iii) How to improve the scalable capability of the clustering algorithm to support large-scale moving trajectory streams? In this paper, we propose an efficient and distributed trajectory stream clustering framework that can: i) model trajectory streams dynamically and effectively in a self-adaptive manner, i.e.,$k$-Segment, which considers both spatial and temporal aspects of trajectory streams, ii) support distributed indexing, processing, and workload balance, and iii) incrementally cluster trajectory streams in an efficient manner. Experiments on a wide range of real-world trajectory datasets show that our framework outperforms state-of-the-art baselines in terms of clustering quality, efficiency, and scalability.

Spatio-temporal trajectory analytics are useful in diversified applications such as urban planning, infrastructure development, and vehicular networks. Trajectory similarity measure, which aims to evaluate the distance between two trajectories, is a fundamental functionality of trajectory analytics. In this paper, we propose a comprehensive survey that investigates all the most common and representative spatio-temporal trajectory measures. First, we provide an overview of spatio-temporal trajectory measures in terms of three hierarchical perspectives: Non-learning versus Learning, Free Space versus Road Network, and Standalone versus Distributed. Next, we present an evaluation benchmark by designing five real-world transformation scenarios. Based on this benchmark, extensive experiments are conducted to study the effectiveness, robustness, efficiency, and scalability of each measure, which offers guidelines for trajectory measure selection among multiple techniques and applications such as trajectory data mining, deep learning, and distributed processing. Specifically, i) Effectiveness: In terms of trajectory length, DFD and Seg-Frechet are length-sensitive, while OWD and Hausdorff always return same results when varying query trajectory length. In terms of trajectory shape, LCRS and LORS are able to effectively find similar trajectories for query trajectories with different shapes; ii) Robustness: Learning based measures are more robust compared with non-learning based ones. Among non-learning based measures, DFD, Hausdorff, OWD and Seg-Frechet are relatively non-sensitive to noises and different sampling rates; and iii) Efficiency& Scalability: Compared to non-learning based measures, learning based and distributed based measures are more efficient and scalable.

The ubiquitous GPS sensors collect massive trajectory data from moving objects, which is useful in data mining applications. However, trajectory data is enormous in volume, and thus, directly storing and processing the raw data is expensive. Using trajectory simplification, a trajectory can be reduced to a set of continuous line segments with acceptable data loss, which is an efficient method. Although many algorithms are proposed, they still suffer from the following issues including (i) non-data driven capability as most studies rely on human-crafted rules or pre-defined parameters, (ii) bound with error measures that yield high computational cost, and (iii) focusing only on the local information preservation in trajectories, but failing in capturing the global mobility patterns for trajectory compression.To address the above issues, we propose a Seq2Seq2Seq framework, abbreviated S3, which consists of two chained Seq2Seq. With differentiable reconstruction learning, S3 enables self-supervised trajectory simplification in a lightweight manner. Besides, we deploy S3 over the graph neural architecture to capture the context-aware mobility patterns and enhance the representation paradigm of trajectories with geographical semantics, where a context-aware distance measure is designed for quality evaluation. An online extension of S3 is also developed to enable streaming trajectory simplifications. Finally, extensive experiments using two real-world datasets in both offline and online scenarios show that S3 achieves much higher efficiency (e.g., it achieves up to one order of magnitude speed-up gains) and comparable compression quality, compared with both non-learning and state-of-the-art learning-based methods.

Federated learning (FL) has emerged as a paradigm to train a global machine learning model in a distributed manner while taking privacy concerns and data protection regulations into consideration. Although a variety of FL algorithms have been proposed, the training efficiency of FL remains challenging due to massive mathematical computations and expensive client-server communication costs. However, existing FL-acceleration studies are limited as they can only solve the computation and communication overheads separately, which is suboptimal and constrains their acceleration ability. Moreover, previous studies are typically designed for specific FL scenarios and can support only one or two FL models, thus exhibiting poor generality.To fill these critical voids, we propose FLBooster, which provides unified and efficient acceleration capacity for a broad range of FL models. This is the first proposal to solve the computation and communication overheads simultaneously. Specifically, we utilize GPUs to boost the computation-intensive homomorphic encryption (HE) operations in a parallel manner, which significantly reduces the computation costs. On the other hand, a simple but efficient compression method is designed to lighten the exchange of data volumes between client and server. Extensive experiments using four standard FL models on three datasets show that FLBooster acquires superior speed-up gains (i.e., 14.3× – 138×) over state-of-the-art acceleration systems. Finally, we integrate FLBooster into the open-source FL benchmark FATE and offer user-friendly APIs for development.

Entity Resolution (ER) is a fundamental problem in data preparation. Standard deep ER methods have achieved state-of-the-art effectiveness, assuming that relations from different organizations are centrally stored. However, due to privacy concerns, it can be difficult to centralize data in practice, rendering standard deep ER solutions inapplicable. Despite efforts to develop rule-based privacy-preserving ER methods, they often neglect subtle matching mechanisms and have poor effectiveness as a result. To bridge effectiveness and privacy, in this paper, we propose CampER, an effective framework for privacy-aware deep entity resolution. Specifically, we first design a training pair self-generation strategy to overcome the absence of manually labeled data in privacy-aware scenarios. Based on the self-constructed training pairs, we present a collaborative fine-tuning approach to learn the match-aware and uni-space individual tuple embeddings for accurate matching decisions. During the matching decision-making process, we first introduce a cryptographically secure approach to determine matches. Furthermore, we propose an order-preserving perturbation strategy to significantly accelerate the matching computation while guaranteeing the consistency of ER results. Extensive experiments on eight widely-used benchmark datasets demonstrate that CampER not only is comparable with the state-of-the-art standard deep ER solutions in effectiveness, but also preserves privacy.

Trajectory similarity queries, including similarity search and similarity join, offer a foundation for many geo-spatial applications. With the rapid increase of streaming trajectory data volumes, e.g., data from mobile phones, vessel monitoring, or traffic systems, many location-based services benefit from online similarity analytics over trajectory data streams, where moving objects continually emit real-time position data. However, most existing studies focus on offline settings, and thus several major challenges remain unanswered in an online setting. To this end, we describe Ghost, a distributed stream processing framework that enables generic, efficient, and scalable online trajectory similarity search and join. We propose a novel incremental online similarity computation (IOSC) mechanism to accelerate pair-wise streaming trajectory distance calculation, which supports a broad range of trajectory distance metrics. Compared with previous studies, IOSC reduces the complexity from quadratic to linear in terms of trajectory length. Building on this foundation, we propose histogram-based algorithms that exploit histogram indexes and a series of pruning bounds to enable streaming trajectory similarity search and join. Finally, we extend our methods to the distributed platform Flink for scalability, where a CostPartitioner is developed to ensure parallel processing and workload balancing. An experimental study using two real-life and one synthetic datasets shows that Ghost (i) acquires 6-20× efficiency/throughput gains and one order of magnitude memory overhead savings over state-of-the-art baselines, (ii) achieves 3--8× workload balancing gains on Flink, and (iii) exhibits low parameter sensitivity and high robustness.

Trajectory data has the potential to greatly benefit a wide-range of real-world applications, such as tracking the spread of the disease through people's movement patterns and providing personalized location-based services based on travel preference. However, privacy concerns and data protection regulations have limited the extent to which this data is shared and utilized. To overcome this challenge, local differential privacy provides a solution by allowing people to share a perturbed version of their data, ensuring privacy as only the data owners have access to the original information. Despite its potential, existing point-based perturbation mechanisms are not suitable for real-world scenarios due to poor utility, dependence on external knowledge, high computational overhead, and vulnerability to attacks. To address these limitations, we introduce LDPTrace, a novel locally differentially private trajectory synthesis framework. Our framework takes into account three crucial patterns inferred from users' trajectories in the local setting, allowing us to synthesize trajectories that closely resemble real ones with minimal computational cost. Additionally, we present a new method for selecting a proper grid granularity without compromising privacy. Our extensive experiments using real-world as well as synthetic data, various utility metrics and attacks, demonstrate the efficacy and efficiency of LDPTrace.

A knowledge graph (KG) consists of a set of interconnected typed entities and their attributes. Recently, KGs are popularly used as the auxiliary information to enable more accurate, explainable, and diverse user preference recommendations. Specifically, existing KG-based recommendation methods target modeling high-order relations/dependencies from long connectivity user-item interactions hidden in KG. However, most of them ignore the cold-start problems (i.e., user cold-start and item cold-start) of recommendation analytics, which restricts their performance in scenarios when involving new users or new items. Inspired by the success of meta-learning on scarce training samples, we propose a novel meta-learning based framework called MetaKG, which encompasses a collaborative-aware meta learner and a knowledge-aware meta learner, to capturemetausers’ preference and entities’ knowledge for cold-start recommendations. The collaborative-aware meta learner aims to locally aggregate user preferences for each preference learning task. In contrast, the knowledge-aware meta learner is to globally generalize knowledge representation across different user preference learning tasks. Guided by two meta learners, MetaKG can effectively capture the high-order collaborative relations and semantic representations, which could be easily adapted to cold-start scenarios. Besides, we devise a novel adaptive task scheduler which can adaptively select the informative tasks for meta learning in order to prevent the model from being corrupted by noisy tasks. Extensive experiments on various cold-start scenarios using three real datasets demonstrate that our presented MetaKG outperforms all the existing state-of-the-art competitors in terms of effectiveness, efficiency, and scalability.

Deep learning based trajectory similarity computation holds the potential for improved efficiency and adaptability over traditional similarity computation. However, existing learning-based trajectory similarity learning solutions prioritize spatial similarity over temporal similarity, making them suboptimal for time-aware analyses. To this end, we propose ST2Vec, a representation learning based solution that considers fine-grained spatial and temporal relations between trajectories to enable spatio-temporal similarity computation in road networks. Specifically, ST2Vec encompasses two steps: (i) spatial and temporal modeling that encode spatial and temporal information of trajectories, where a generic temporal modeling module is proposed for the first time; and (ii) spatio-temporal co-attention fusion, where two fusion strategies are designed to enable the generation of unified spatio-temporal embeddings of trajectories. Further, under the guidance of triplet loss, ST2Vec employs curriculum learning in model optimization to improve convergence and effectiveness. An experimental study offers evidence that ST2Vec outperforms state-of-the-art competitors substantially in terms of effectiveness and efficiency, while showing low parameter sensitivity and good model robustness. Moreover, similarity involved case studies including top-k querying and DBSCAN clustering offer further insight into the capabilities of ST2Vec.

The ubiquity of implicit feedback makes them the default choice to build modern recommender systems. Generally speaking, observed interactions are considered as positive samples, while unobserved interactions are considered as negative ones. However, implicit feedback is inherently noisy because of the ubiquitous presence of noisy-positive and noisy-negative interactions. Recently, some studies have noticed the importance of denoising implicit feedback for recommendations, and enhanced the robustness of recommendation models to some extent. Nonetheless, they typically fail to (1) capture the hard yet clean interactions for learning comprehensive user preference, and (2) provide a universal denoising solution that can be applied to various kinds of recommendation models.

Trajectory clustering has played an essential role in trajectory mining tasks. It serves in a wide range of real-life applications, including transportation, location-based services, behavioral study, and so on. To support trajectory clustering analytics, a plethora of trajectory clustering methods have been proposed, which mainly extend traditional clustering algorithms by using spatio-temporal characteristics of trajectories. However, existing traditional trajectory clustering approaches based on raw trajectory representation highly rely on hand-craft similarity metrics, and can not capture hidden spatial dependencies in trajectory data, which is inefficient and inflexible for clustering analysis. To this end, we propose an end-to-end deep trajectory clustering framework via self-training, termed as E2DTC, inspired by the data-driven capabilities of deep neural networks. E2DTC does not require any additional manual feature extraction operations, and can be easily adapted for trajectory clustering analytics on any trajectory dataset. Extensive experimental evaluations on three real-life datasets show that our framework E2DTC achieves superior accuracy and efficiency, compared with classical clustering methods (i.e., K-Medoids) and state-of-the-art neural-network based approaches (i.e., t2vec).

Traffic prediction has drawn increasing attention for its ubiquitous real-life applications in traffic management, urban computing, public safety, and so on. Recently, the availability of massive trajectory data and the success of deep learning motivate a plethora of deep traffic prediction studies. However, the existing neural-network-based approaches tend to ignore the correlations between multiple types of moving objects located in the same spatio-temporal traffic area, which is suboptimal for traffic prediction analytics. In this paper, we propose a multi-source deep traffic prediction framework over spatio-temporal trajectory data, termed as MDTP. The framework includes two phases: spatio-temporal feature modeling and multi-source bridging. We present an enhanced graph convolutional network (GCN) model combined with long short-term memory network (LSTM) to capture the spatial dependencies and temporal dynamics of traffic in the feature modeling phase. In the multi-source bridging phase, we propose two methods, Sum and Concat, to connect the learned features from different trajectory data sources. Extensive experiments on two real-life datasets show that MDTP i) has superior efficiency, compared with classical time-series methods, machine learning methods, and state-of-the-art neural-network-based approaches; ii) offers a significant performance improvement over the single-source traffic prediction approach; and iii) performs traffic predictions in seconds even on tens of millions of trajectory data. we develop MDTP + , a user-friendly interactive system to demonstrate traffic prediction analysis.

The aim of real-time co-movement pattern mining for streaming trajectories is to discover co-moving objects that satisfy specific spatio-temporal constraints in real time. This functionality serves a range of real-world applications, such as traffic monitoring and management. However, little work targets the visualization and interaction with such co-movement detection on streaming trajectories. To this end, we develop CoMing, a real-time co-movement pattern mining system, to handle streaming trajectories. CoMing leverages ICPE, a real-time distributed co-movement pattern detection framework, and thus, it has its capacity of good performance. This demonstration offers hands-on experience with CoMing's visual and user-friendly interface. Moreover, several applications in the traffic domain, including object monitoring and traffic statistics visualization, are also provided to users.

With the widespread deployment of mobile devices with positioning capabilities, increasingly massive volumes of trajectory data are being collected that capture the movements of people and vehicles. This data enables co-movement pattern detection, which is important in applications such as trajectory compression and future-movement prediction. Existing co-movement pattern detection studies generally consider historical data and thus propose offline algorithms. However, applications such as future movement prediction need real-time processing over streaming trajectories. Thus, we investigate real-time distributed co-movement pattern detection over streaming trajectories. Existing off-line methods assume that all data is available when the processing starts. Nevertheless, in a streaming setting, unbounded data arrives in real time, making pattern detection challenging. To this end, we propose a framework based on Apache Flink, which is designed for efficient distributed streaming data processing. The framework encompasses two phases: clustering and pattern enumeration. To accelerate the clustering, we use a range join based on two-layer indexing, and provide techniques that eliminate unnecessary verifications. To perform pattern enumeration efficiently, we present two methods FBA and VBA that utilize id-based partitioning. When coupled with bit compression and candidate-based enumeration techniques, we reduce the enumeration cost from exponential to linear. Extensive experiments offer insight into the efficiency of the proposed framework and its constituent techniques compared with existing methods.