Significance Identification and quantification of influential spreaders in social networks are challenging due to the gigantic network sizes limited availability entire structure. Here we show that such difficulty can be overcome by reducing problem scale a local one, which is essentially independent network. This because viral spreading characteristic size does not depend on structure outside environment seed spreaders. Our approach may open door solve various big data problems as false...

Urban metro flow prediction is of great value for operation scheduling, passenger management and personal travel planning. However, the problem challenging. First, different stations, e.g. transfer stations non-transfer have unique traffic patterns. Second, it difficult to model complex spatio-temporal dynamic relation stations. To address these challenges, we develop a graph relational learning (STDGRL) predict urban station flow. propose node embedding representation module capture...

Urban sensing is a foundation of urban computing, collecting data in cities through ubiquitous computing techniques, e.g. using humans as sensors. In this paper, we propose crowd-based framework that maximizes the coverage collected spatio-temporal space, based on human mobility participants recruited by given budget. This provides with unobstructed tasks do not break their original commuting plans, while ensuring program balanced better supports upper-level applications. The consists three...

Emergency medical service provides a variety of services for those in need emergency care. One the major challenges encountered by providers is selecting appropriate locations ambulance stations. Prior works measure spatial proximity under Euclidean space or static road network. In this paper, we focus on locating stations using real traffic information so as to minimize average travel-time reach requests. To end, estimate segments GPS trajectories and propose an efficient PAM-based...

Urban flow analysis is an essential research for smart city construction, in which urban pattern focuses on the continuous state of flow. How to mine, store and reuse traffic patterns from multi-source heterogeneous big data challenging. Therefore, this paper proposes a knowledge mining network regional mine pattern. The proposed model consists two modules. In first module, features region its are extracted as entity relation, respectively. second POI modeled enhance embedding representation...

Recommender system can provide users with the required information accurately and efficiently, playing a very important role in improving users' life experience. Although knowledge graph-based recommender solve sparsity cold start problems faced by traditional system, it cannot handle cross-domain problem multi-domain recommendations. Therefore, this paper focuses on item-item (I2I) recommendation based graph embedding analyzing association between items of same domain interaction diverse...

Protecting citizens' lives from emergent accidents (e.g. traffic accidents) and diseases heart attack) is of vital importance in urban computing. Every day many people are caught or thus need ambulances to transport them hospitals. In this paper, we propose a dynamic ambulance redeployment system reduce the time needed for pick up patients increase probability being saved time. For danger, every second counts. Specifically, whenever there an becoming available finishing transporting patient...

Ordering take-out food (a.k.a. takeaway food) on online-to-offline (O2O) ordering and delivery platforms is becoming a new lifestyle for people living in big cities, thanks to its great convenience. Web users mobile device can order (i.e. obtain online services) an O2O platform. Then the platform will dispatch carriers deliver from restaurants users, i.e. providing with offline services. For platform, improving efficiency, given massive number of orders each day limited carriers, paramount...

Daily schedule recommendation is an intelligent approach to recommend multiple suitable activity locations and sequences for users based on their needs in a day. In such scenario, training the model using traditional methods requires centralized data collection from individual users, which may be prohibited by protection acts, as GDPR CCPA. this paper, we address problem of daily utilizing deep reinforcement learning federated framework (FedDSR). And curriculum applied guide process towards...

Social networks constitute a new platform for information propagation, but its success is crucially dependent on the choice of spreaders who initiate spreading information. In this paper, we remove edges in network at random and segments into isolated clusters. The most important nodes each cluster then form set influential spreaders, such that news propagating from them would lead to extensive coverage minimal redundancy. method utilizes similarities between segmented before percolation...

Bootstrap percolation is a well-known model to study the spreading of rumors, new products or innovations on social networks. The empirical studies show that community structure ubiquitous among various Thus, studying bootstrap complex networks with communities can bring us and important insights into dynamics in This has attracted lot scientists' attention recently. In this letter, we Erdős-Rényi observed second-order, hybrid (both second first order) multiple phase transitions, which rare...

Representing the structural relations between entities, i.e., knowledge graph embedding, which is a method to learn low-dimensional representations of knowledge, has become an increasingly prevalent research orientation in cognitive and human intelligence. It significant study how interrelate, fuse embed data from different domains while considering not shared. In this paper, we propose model cross-domain embedding federated learning (FedCKE), entity/relation can interact securely case that...

In our daily lives, people frequently consider schedule to meet their needs, such as going a barbershop for haircut, then eating in restaurant, and finally shopping supermarket. Reasonable activity location or point-of-interest (POI) sequencing will help save lot of time get better services. this article, we propose reinforcement learning-based deep factor balancing model recommend reasonable according user's current needs. The proposed consists network (DAFB) learning framework. First, the...

Fusion technique is a key research topic in multimodal sentiment analysis. The recent attention-based fusion demonstrates advances over simple operation-based fusion. However, these works adopt single-scale, i.e., token-level or utterance-level, unimodal representation. Such single-scale suboptimal because that different modality should be aligned with granularities. This paper proposes model named ScaleVLAD to gather multi-Scale representation from text, video, and audio shared Vectors of...

Emergency Medical Services (EMS) are of great importance to saving people's lives from emergent accidents and diseases by efficiently picking up patients using ambulances. The transporting capability an EMS system (e.g., defined as the average pickup time patients) significantly depends on real-time redeployment strategy That is, which station should ambulance be redeployed to, after it becomes available (after transports a patient hospital or finishes in-site treatment for patient)?...

Social networks constitute a new platform for information propagation, but its success is crucially dependent on the choice of spreaders who initiate spreading information. In this paper, we remove edges in network at random and segments into isolated clusters. The most important nodes each cluster then form group influential spreaders, such that news propagating from them would lead to an extensive coverage minimal redundancy. method well utilizes similarities between pre-percolated state...

As travel efficiency matters to the work productivity of cities, shortening passengers' time for metros is therefore a pressing need. To this end, we study strategy by dynamically scheduling dwell trains. Developing such challenging because three aspects: 1) Optimizing average passengers needs properly balance waiting at platforms and journey on trains, as well considering long-term impacts; 2) Capturing dynamic spatio-temporal (ST) correlations incoming metro stations difficult; 3) For each...