Abstract: UASNs are used in offshore exploration, auxiliary navigation, and marine military. Underwater sensor nodes’ communication, computation, and storage limitations preclude traditional security mechanisms. Trust models have been investigated as effective tools for UASN security…
Abstract: Recent smartphone sensors make identifying a locomotion mode easy. Locomomotion mode information improves trip planning, travel time estimation, and traffic management. Locomotion mode recognition work depends on labeled training instances and is not relevant….
Abstract: Network-based indoor positioning with WiFi access point (AP) RSS measurements uses channel modeling, position estimation, and error analysis. A novel sparse Bayesian learning algorithm models the radio power map (RPM) in indoor space to…
Abstract: Indoor positioning using Wi-Fi fingerprinting is common. It uses a similarity function to compare the operational fingerprint to a radio map of reference samples. In large deployments with a high fingerprint density and a…
Abstract: Unsupervised spectral feature selection (USFS) methods can produce interpretable and discriminative results by embedding a Laplacian regularizer in sparse feature selection to maintain training sample local similarity. USFS methods use a general-graph or hyper-graph…
Abstract: Social neighborhoods influence people’s social decisions. Structural diversity in the contact neighborhood predicts social contagion. Existing models have limited decomposability for large-scale networks and inaccurately reflect social context diversity. This paper proposes a truss-based…
Abstract: We study an automatic hypothesis generation (HG) problem, which involves discovering meaningful implicit connections between scientific terms like diseases, chemicals, drugs, and genes extracted from biomedical publication databases. Most previous studies of this problem…
Abstract: Recently, social media research has grown. This article predicts active users in online social events, which is rarely done. Previous work focused on emerging topics. This prediction task is a binary classification problem based…
Abstract: CNNs automatically extract high-level features from remote sensing (RS) images. PCANet, CCANet, MS-CCANet, and MCCANet are CNN model variants. RS data are often observed from multiple views, but the PCANet is specialized for single-view…
Abstract: Subgraph matching finds all data graph embeddings that match a query graph. Recent algorithms create a tree-structured index on the data graph based on the query graph, order the vertices path-by-path, and enumerate the…
Abstract: We propose Stat-DSM to assess the statistical significance (reliability) of discriminative sub-trajectory mining results. Stat-DSM extracts sub-trajectories that occur statistically significantly more often in one group than the other from two groups of trajectories….
Abstract: Predicting urban traffic is crucial to intelligent transportation systems and public safety, but it is difficult due to three factors: 1) complex spatio-temporal correlations of urban traffic, including spatial correlations between locations and temporal…
WhatsApp us