Complex user behavior, especially in settings such as social media, can be organized as time-evolving networks. Through network embedding, we can extract general-purpose vector representations of these dynamic networks which allow us to analyze them without extensive feature engineering. Prior work has shown how to generate network embeddings while preserving the structural role proximity of nodes. These methods, however, cannot capture the temporal evolution of the structural identity of the nodes in dynamic networks. Other works, on the other hand, have focused on learning microscopic dynamic embeddings. Though these methods can learn node representations over dynamic networks, these representations capture the local context of nodes and do not learn the structural roles of nodes. In this article, we propose a novel method for learning structural node embeddings in discrete-time dynamic networks. Our method, called
, tracks historical topology information in dynamic networks to learn dynamic structural role embeddings. Through experiments on synthetic and real-world temporal datasets, we show that our method outperforms other well-known methods in tasks where structural equivalence and historical information both play important roles.
can be used to model dynamic user behavior in any networked setting where users can be represented as nodes. Additionally, we propose a novel method (called network fingerprinting) that uses
embeddings for modeling whole (or partial) time-evolving networks. We showcase our network fingerprinting method on synthetic and real-world networks. Specifically, we demonstrate how our method can be used for detecting foreign-backed information operations on Twitter.
IoT and the Cloud are among the most disruptive changes in the way we use data today. These changes have not significantly influenced practices in condition monitoring for shipping. This is partly due to the cost of continuous data transmission. Several vessels are already equipped with a network of sensors. However, continuous monitoring is often not utilised and onshore visibility is obscured. Edge computing is a promising solution but there is a challenge sustaining the required accuracy for predictive maintenance. We investigate the use of IoT systems and Edge computing, evaluating the impact of the proposed solution on the decision making process. Data from a sensor and the NASA-IMS open repository were used to show the effectiveness of the proposed system and to evaluate it in a realistic maritime application. The results demonstrate our real-time dynamic intelligent reduction of transmitted data volume by
without sacrificing specificity or sensitivity in decision making. The output of the Decision Support System fully corresponds to the monitored system's actual operating condition and the output when the raw data are used instead. The results demonstrate that the proposed more efficient approach is just as effective for the decision making process.
Since the outbreak of the coronavirus disease 2019 (COVID-19) pandemic, the pandemic situation has begun to undergo positive changes with the joint efforts of various countries and world organizations. However, pressures such as the COVID-19 mutations and the sharp rise in confirmed cases have brought uncertainties to the prevention and control of the pandemic. The overall situation is still severe and complex. Based on the multi-dimensional spatial-temporal COVID-19 data collected by the open-source NetEase News (NEN) website and a real-time dynamic website, it is to explore the characteristics of the pandemic data, visualize the development trend, and analyze the spread of the pandemic in this paper. Moreover, it is to provide a rule basis for the prevention and control of the COVID-19 pandemic by constructing the decision tree model. From the results, some suggestions are provided for decision-makers.
AbstractLocoregional therapies (LRTs) are an essential management tool in the treatment of primary liver cancers or metastatic liver disease. LRTs include curative and palliative modalities. Monitoring treatment response of LRTs is crucial for maximizing benefit and improving clinical outcomes. Clinical use of contrast-enhanced ultrasound (CEUS) was introduced more than two decades ago. Its portability, cost effectiveness, lack of contraindications and safety make it an ideal tool for treatment monitoring in numerous situations. Two-dimensional dynamic CEUS has been proved to be equivalent to the current imaging standard in the guidance of LRTs, assessment of their adequacy, and detection of early tumor recurrence. Recent technical advances in ultrasound transducers and image processing have made 3D CEUS scanning widely available on most commercial ultrasound systems. 3D scanning offers a broad multiplanar view of anatomic structures, overcoming many limitations of two-dimensional scanning. Furthermore, many ultrasound systems provide real-time dynamic 3D CEUS, also known as 4D CEUS. Volumetric CEUS has shown to perform better than 2D CEUS in the assessment and monitoring of some LRTs. CEUS presents a valid alternative to the current imaging standards with reduced cost and decreased risk of complications. Future efforts will be directed toward refining the utility of 4D CEUS through approaches such as multi-parametric quantitative analysis and machine learning algorithms.
Dynamic window algorithm (DWA) is a local path-planning algorithm, which can be used for obstacle avoidance through speed selection and obtain the optimal path, but the algorithm mainly plans the path for fixed obstacles. Based on DWA algorithm, this paper proposes an improved DWA algorithm based on space-time correlation, namely, space-time dynamic window approach. In SDWA algorithm, a DWA associated with obstacle position and time is proposed to achieve the purpose of path planning for moving obstacles. Then, by setting the coordinates of the initial moving obstacle and identifying safety distance, we can define the shape of the obstacle and the path planning of the approach segment in thunderstorm weather based on the SDWA model was realized. Finally, the superior performance of the model was verified by setting moving obstacles for path planning and selecting the aircraft approach segment in actual thunderstorm weather. The results showed that SDWA has good path-planning performance in a dynamic environment. Its path-planning results were very similar to an actual aircraft performing thunderstorm-avoidance maneuvers, but with more smooth and economical trajectory. The proposed SDWA model had great decision-making potential for approach segment planning in thunderstorm weather.
We report a universal phase reconfiguration phenomenon and a doping strategy to enhance the activity of multivalent nickel sulfides in hydrogen evolution. Based on these, a life-time dynamic structure-activity correlation has been established.
We highlight the development of activatable molecular probes that trigger the optical signals toward biomarkers, allowing real-time, dynamic visualization of lesions and margins for guided-surgery, endoscopy and tissue biopsy with molecular precision.
The range of values of the coefficient of resistance to movement of the chain of typical longwall armored face conveyors and the coefficient of inner viscous friction in the chain, both immersed in the moving load and during the idle run of the conveyor, is estimated. The computer model of the conveyor is built as a multi-mass elastic-viscous stretched closed chain without sag with the number of masses n = 200 and one induction drive motor located in the head of the conveyor. Using the constructed model, three-dimensional space-time dynamic characteristics of speeds and forces in the chain of the CP72 longwall armored face conveyor are obtained. Start up to rated speed v≈1 m / s and the working process is simulated with an unloaded conveyor. The spatial form of frictional self-oscillations in the model with distributed parameters is shown. The resonance frequencies and amplitudes of oscillations of the efforts in the circuit and the length of the corresponding spatial waves have been determined. It was found that at the first and second resonance frequencies, self-oscillations are not excited, since the damping effect of the electric drive is quite pronounced in this frequency band. The direct connection of vibration amplitudes with the energy efficiency of the conveyor electric drive is indicated.