Recently, spatial keyword queries become a hot topic in the literature. One example of these is collective query (CoSKQ) which to find set objects database such that it covers given keywords collectively and has smallest cost. Unfortunately, existing exact algorithms have severe scalability problems approximate algorithms, though scalable, cannot guarantee near-to-optimal solutions. In this paper, we study CoSKQ problem address above issues.

Trajectories of moving objects are collected in many applications. Raw trajectory data is typically very large, and has to be simplified before use. In this paper, we introduce the notion direction-preserving simplification, show both analytically empirically that it can support a broader range applications than traditional position-preserving simplification. We present polynomial-time algorithm for optimal another approximate with quality guarantee. Extensive experimental evaluation real...

Online bipartite graph matching is attracting growing research attention due to the development of dynamic task assignment in sharing economy applications, where tasks need be assigned dynamically workers. Past studies lack practicability terms both problem formulation and solution framework. On one hand, some settings prior online are impractical for real-world applications. other existing solutions inefficient unnecessary real-time decision making. In this paper, we propose (DBGM) better...

Abstract An increasing number of terrestrial‐ and space‐based radio‐communication systems are influenced by the ionospheric space weather, making state increasingly important to forecast. In this study, a novel extended encoder‐decoder long short‐term memory (ED‐LSTME) neural network, which can predict total electron content (TEC) is proposed. Useful inherent features were automatically extracted from historical TEC LSTM layers, performance proposed model was enhanced considering solar flux...

Learned indexes have been proposed to replace classic index structures like B-Tree with machine learning (ML) models. They require both the and query processing algorithms currently deployed by databases, such a radical departure is likely encounter challenges obstacles. In contrast, we propose fundamentally different way of using ML techniques build better R-Tree without need change structure or traditional R-Tree. Specifically, develop reinforcement (RL) based models decide how choose...

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...

Mobility prediction, which is to predict where a user will arrive based on the user's historical mobility records, has attracted much attention. We argue that it more useful know not only but also when next in many scenarios such as targeted advertising and taxi service. In this paper, we propose novel context-aware deep model called DeepJMT for jointly performing prediction (to where) time when). The consists of (1) hierarchical recurrent neural network (RNN) sequential dependency encoder,...

Fake news has become more prevalent than ever, correlating with the rise of social media that allows every user to rapidly publish their views or hearsay. Today, fake spans almost realm human activity, across diverse fields such as politics and healthcare. Most existing methods for detection leverage supervised learning expect a large labelled corpus articles engagement information, which are often hard, time-consuming costly procure. In this paper, we consider task unsupervised detection,...

Approximate K nearest neighbor (AKNN) search in the high-dimensional Euclidean vector space is a fundamental and challenging problem. We observe that space, time consumption of nearly all AKNN algorithms dominated by distance comparison operations (DCOs). For each operation, it scans full dimensions an object thus, runs linear wrt dimensionality. To speed up, we propose randomized algorithm named ADSampling which logarithmic dimensionality for majority DCOs succeeds with high probability. In...

Searching for approximate nearest neighbors (ANN) in the high-dimensional Euclidean space is a pivotal problem. Recently, with help of fast SIMD-based implementations, Product Quantization (PQ) and its variants can often efficiently accurately estimate distances between vectors have achieved great success in-memory ANN search. Despite their empirical success, we note that these methods do not theoretical error bound are observed to fail disastrously on some real-world datasets. Motivated by...

Viral marketing has attracted considerable concerns in recent years due to its novel idea of leveraging the social network propagate awareness products. Specifically, viral is first target a limited number users (seeds) by providing incentives, and these targeted would then initiate process spread propagating information their friends via relationships. Extensive studies have been conducted for maximizing given seeds. However, all them fail consider common scenario where companies hope use...

Trajectory data is central to many applications with moving objects. Raw trajectory usually very large, and so simplified before it stored processed. Many simplification notions have been proposed, among them, the direction-preserving (DPTS) which aims at protecting direction information has shown perform quite well. However, existing studies on DPTS require users specify an error tolerance might not know how set properly in some cases (e.g., could only be known future time simply setting...

In this paper, we study the optimal location query problem based on road networks. Specifically, have a network which some clients and servers are located. Each client finds server that is closest to her for service cost of getting served equal (network) distance between serving multiplied by weight or importance. The find setting up new such maximum being (including server) minimized. This has been studied before, but state-of-the-art still not efficient enough. propose an algorithm...

Extracting interesting tuples from a large database is an important problem in multi-criteria decision making. Two representative queries were proposed the literature: top- k and skyline queries. A query requires users to specify their utility functions beforehand then returns users. does not require any function but it puts no control on number of returned Recently, k-regret was received attention community because output size controllable, thus avoids those deficiencies Specifically, that...

In building-microgrid communities, renewable generation and time-varying load usually cause power fluctuations, which influence the ancillary support to main grid. Thermostatically controlled loads (TCLs) can be utilized compensate such variations due their aggregated controllable consumptions. Meanwhile, one basic requirement for users' side of TCLs is realize fair sharing states comfort states. This article proposes a distributed event-based control strategy, where information neighboring...

In this work, we propose a robust road network representation learning framework called Toast, which comes to be cornerstone boost the performance of numerous demanding transport planning tasks. Specifically, first traffic context aware skip-gram module incorporate auxiliary tasks predicting target segment. Furthermore, trajectory-enhanced Transformer that utilizes trajectory data extract traveling semantics on networks. Apart from obtaining effective segment representations, also enables us...

Road extraction is a process of automatically generating road maps mainly from satellite images. Existing models all target to generate roads the scratch despite that large quantity maps, though incomplete, are publicly available (e.g. those OpenStreetMap) and can help with extraction. In this paper, we propose conduct based on images partial which new. We then two-branch Partial Complete Network (P2CNet) for task, has two prominent components: Gated Self-Attention Module (GSAM) Missing Part...

The robustness of recommender systems has become a prominent topic within the research community. Numerous adversarial attacks have been proposed, but most them rely on extensive prior knowledge, such as all white-box or black-box which assume that certain external knowledge is available. Among these attacks, model extraction attack stands out promising and practical method, involving training surrogate by repeatedly querying target model. However, there significant gap in existing...

Abstract To solve the problem of high impedance line‐to‐ground fault, a solution using isolation transformers for subnetwork divisions was previously implemented in power distribution network. As result that, network has been experiencing ferroresonance more often. Through modelling and simulation based on its actual parameters PSCAD/EMTDC, understanding situation is deepened. The originates from response to fault clearing. division certainly improved damping factor, but line capacitance...

Listing k-cliques plays a fundamental role in various data mining tasks, such as community detection and of cohesive substructures. Existing algorithms for the k-clique listing problem are built upon general framework, which finds by recursively finding (k-1)-cliques within subgraphs induced out-neighbors each vertex. However, this framework has inherent inefficiency smaller cliques certain repeatedly. In paper, we propose an algorithm DIST problem. contrast to existing works, main idea our...