Cycling as a green transportation mode has been promoted by many governments all over the world. As result, constructing effective bike lanes become crucial task for promoting cycling life style, well-planned paths can reduce traffic congestion and decrease safety risks both cyclists motor vehicle drivers. Unfortunately, existing trajectory mining approaches lane planning do not consider key realistic government constraints: 1) budget limitations, 2) construction convenience, 3) utilization.

With the prevalence of positioning techniques, a prodigious number spatio-temporal data is generated constantly. To effectively support sophisticated urban applications, e.g., location-based services, based on data, it desirable for an efficient, scalable, update-enabled, and easy-to-use management system.This paper presents JUST, i.e., JD Urban Spatio-Temporal engine, which can efficiently manage big in convenient way. JUST incorporates distributed NoSQL store, Apache HBase, as underlying...

Accurate and updated road network data is vital in many urban applications, such as car-sharing, logistics. The traditional approach to identifying the network, i.e., field survey, requires a significant amount of time effort. With wide usage GPS embedded devices, huge trajectory has been generated by different types mobile objects, which provides new opportunity extract underlying network. However, existing trajectory-based map recovery approaches require empirical parameters do not utilize...

Trajectory data is very useful for many urban applications. However, due to its spatio-temporal and high-volume properties, it challenging manage trajectory data. Existing management frameworks suffer from scalability problem, only support limited queries. This paper proposes a holistic distributed NoSQL storage engine, TrajMesa, based on GeoMesa, an open-source indexing toolkit TrajMesa adopts novel schema, which reduces the size tremendously. We also devise key designs, propose bunch of...

With the increasing adoption of GPS modules, there are a wide range urban applications based on trajectory data analysis, such as vehicle navigation, travel time estimation, and driver behavior analysis. The effectiveness relies greatly high sampling rates trajectories precisely matched to map. However, large number collected under low rate in real-world practice, due certain communication loss energy constraints. To enhance support more effectively, many recovery methods proposed infer free...

Illegal vehicle parking is a common urban problem faced by major cities in the world, as it incurs traffic jams, which lead to air pollution and accidents. Traditional approaches detect illegal events rely highly on active human efforts, e.g., police patrols or surveillance cameras. However, these are extremely ineffective cover large city. The massive high quality sharing bike trajectories from Mobike offer us with unique opportunity design ubiquitous detection system, most of happen at...

Urban flow monitoring systems play important roles in smart city efforts around the world. However, ubiquitous deployment of devices, such as CCTVs, induces a long-lasting and enormous cost for maintenance operation. This suggests need technology that can reduce number deployed while preventing degeneration data accuracy granularity. In this paper, we aim to infer real-time fine-grained crowd flows throughout based on coarse-grained observations. task is challenging due two reasons: spatial...

Spatially fine-grained urban flow data is critical for smart city efforts. Though information desirable applications, it demands much more resources the underlying storage system compared to coarse-grained data. To bridge gap between efficiency and utility, in this paper, we aim infer flows throughout a from their counterparts. This task exhibits two challenges: spatial correlations coarse- flows, complexities of external impacts. tackle these issues, develop model entitled UrbanFM which...

Cross-domain sequential recommendation (CDSR) utilizes data from multiple domains to recommend the user’s next interaction based on his latest sequence. Currently, many cross-domain algorithms have been proven achieve good performance. However, these overlook influence of users’ long-term behavioral patterns and general interests when extracting their current preferences. In this paper, we propose a Hierarchical Gating Network for Cross-Domain Sequential Recommendation (HGNCDSR)....

With the advances of web-of-things, human mobility, e.g., GPS trajectories vehicles, sharing bikes, and mobile devices, reflects people’s travel patterns preferences, which are especially crucial for urban applications such as planning business location selection. However, collecting a large set mobility data is not easy because privacy commercial concerns, well high cost to deploy sensors long time collect data, in newly developed cities. Realizing this, this paper, based on intuition that...

The rapid development of e-commerce requires efficient and reliable logistics services. Nowadays, couriers are still the main solution to address "last mile" problem in logistics. They usually required record accurate delivery time each parcel manually, which provides vital information for applications like insurances, performance evaluations, customer available discovery. Couriers' trajectories generated by their PDAs provide a chance infer automatically ease burdens on couriers. However,...

The objective of public resource allocation, e.g., the deployment billboards, surveillance cameras, base stations, trash bins, is to serve more people. However, due dynamics human mobility patterns, people are distributed unevenly on spatial and temporal domains. As a result, in many cases, redundant resources have be deployed meet crowd coverage requirements, which leads high costs low usage. Fortunately, with development unmanned vehicles, dynamic allocation those becomes possible. To this...

Trajectory data preprocessing is to convert raw GPS logs into organized trajectories, which a common, necessary but tedious task in many urban applications. This paper proposes CloudTP, cloud-based flexible trajectory framework, provide an efficient online service, easing the burdens of application builders. The proposed system designed and implemented based on cloud storage parallel computing framework (i.e. Spark). Its features consist 1) noise filtering, 2) segmentation, 3) map matching,...

The updated residential-level fine-grained digital map is essential for last-mile delivery. However, many of those low-level roads are not recorded in maps due to the high mapping costs. With digitization logistics industry, couriers' trajectories become a promising data source complete missing maps. Existing trajectory-based updating work rely on heavy parameter tuning overcome positioning error issue their unsupervised nature, and able handle issues unreliable road indicators skewed...

Cycling as a green transportation mode has been promoted by many governments all over the world. As result, constructing effective bike lanes become crucial task to promote cycling life style, well-planned can reduce traffic congestions and safety risks. Unfortunately, existing trajectory mining approaches for lane planning do not consider one or more key realistic government constraints: 1) budget limitations, 2) construction convenience, 3) utilization. In this paper, we propose...

With the development of positioning technology, a large number trajectories have been generated, which are very useful for many urban applications. However, it is challenging to manage trajectory data its spatio-temporal dynamics and high-volume properties. Existing management frameworks suffer from efficiency or scalability problem, only support limited query types. This paper takes first attempt build holistic distributed NoSQL storage engine, named TrajMesa, based on GeoMesa, an...

Service time is a part of cost in the last-mile delivery, which spent on delivering parcels at certain location. Predicting service fundamental for many downstream logistics applications, e.g., route planning with windows, courier workload balancing and delivery prediction. Nevertheless, it non-trivial given complex circumstances, location heterogeneity, skewed observations space. The existing solution trains supervised model based aggregated features extracted from to deliver, cannot handle...

Delivery locations are fundamental data source for intelligent logistics, which can be used in route planning, arrival time estimation, parcel allocation, etc. Using the Geocoded way-bill location of an address as delivery is not sufficient, due to wrong parsing, coarse-grained POI database, or different preferences customers. To mitigate insufficiency Geocoding, some methods have been proposed, utilize couriers' when waybills confirmed delivered inference. Nevertheless, these highly rely on...

With the rapid development of location-acquisition techniques, massive trajectories are continuously generated. Many urban applications rely heavily on data mining/analysis results trajectory data. This demo presents a holistic management system for both historical and real-time records based cloud platform, such as Microsoft Azure. The proposed is able to efficiently support variety queries, including ID-Temporal query, Spatio-Temporal Path-Temporal query. these we demonstrate that...

Similarity search has recently become an integral part of many trajectory data analysis tasks. As the number trajectories increases, we must find similar among massive trajectories, necessitating a scalable and efficient frame-work. Typically, can be managed by key-value stores. However, existing works with stores use coarse representation to store data. Besides, they do not provide query processing trajectories. Thus, this paper proposes TraSS, framework for similarity in We propose novel...

Trajectory modeling refers to characterizing human movement behavior, serving as a pivotal step in understanding mobility patterns. Nevertheless, existing studies typically ignore the confounding effects of geospatial context, leading acquisition spurious correlations and limited generalization capabilities. To bridge this gap, we initially formulate Structural Causal Model (SCM) decipher trajectory representation learning process from causal perspective. Building upon SCM, further present...

A path query aims to find trajectories passing a given sequence of connected road segments within time period. It is very useful in many urban applications: 1) traffic modeling, 2) frequent mining, 3) intersection coordination, and 4) anomaly detection. Existing solutions for processing are implemented based on single machines, which not efficient the following tasks: indexing large-scale historical data; handling real-time trajectory updates; concurrent queries from data mining...