It was determined that institutions, public and private worldwide have problems in confidentiality, integrity and authentication in information;due to this it is necessary to look for other alternatives to mitigate th...
详细信息
Graph Neural Networks (GNNs) have achieved remarkable performance in the task of the semi-supervised node classification. However, most existing models learn a deterministic classification function, which lack suffici...
ISBN:
(纸本)9781713829546
Graph Neural Networks (GNNs) have achieved remarkable performance in the task of the semi-supervised node classification. However, most existing models learn a deterministic classification function, which lack sufficient flexibility to explore better choices in the presence of kinds of imperfect observed data such as the scarce labeled nodes and noisy graph structure. To improve the rigidness and inflexibility of deterministic classification functions, this paper proposes a novel framework named Graph Stochastic Neural Networks (GSNN), which aims to model the uncertainty of the classification function by simultaneously learning a family of functions, i.e., a stochastic function. Specifically, we introduce a learnable graph neural network coupled with a high-dimensional latent variable to model the distribution of the classification function, and further adopt the amortised variational inference to approximate the intractable joint posterior for missing labels and the latent variable. By maximizing the lower-bound of the likelihood for observed node labels, the instantiated models can be trained in an end-to-end manner effectively. Extensive experiments on three real-world datasets show that GSNN achieves substantial performance gain in different scenarios compared with state-of-the-art baselines.
Twitter is a widespread supply for real-time news distribution between individuals. Furthermore, spammers could post any kinds of spam content to users, and a variant of incidents are committed on Twitter against user...
Twitter is a widespread supply for real-time news distribution between individuals. Furthermore, spammers could post any kinds of spam content to users, and a variant of incidents are committed on Twitter against users. These threats aren't restricted to the social media platforms however they threaten the safety of Twitter users. Most of the researches use deep learning techniques to detect Twitter spammer activities. The traditional solutions check the behavior of each account or campaign of similar purpose accounts. The number of solutions concentrate on detecting spam campaign based on URL only and ignoring text in a tweet. In this paper, the lightweight framework is proposed to take tweet text into consideration for optimizing spam campaign detection methods based on deep learning techniques. The main contribution of this work summarized in two key points. First one is to summarize text of the tweets to cluster them. The second one is to find similar tweets based on Siamese Recurrent Network. Experimental results show the ability of the proposed technique to extract the right candidate campaign and classify them as spam or not with high recall and precision.
This paper offers new mathematical models to measure the most productive scale size (MPSS) of production systems with mixed structure networks (mixed of series and parallel). In the first property, we deal with a gene...
详细信息
Avoidable hospital readmission is problematic as it increases the burden on healthcare systems, leads to a shortage of hospital beds and impacts on the costs of healthcare. Various machine learning algorithms have bee...
详细信息
ISBN:
(数字)9781728125473
ISBN:
(纸本)9781728125480
Avoidable hospital readmission is problematic as it increases the burden on healthcare systems, leads to a shortage of hospital beds and impacts on the costs of healthcare. Various machine learning algorithms have been applied to predict patient readmissions. However, existing literature has only focused on individual features of health conditions without consideration of associations between features. This paper proposes discriminative pattern-based features as a technique to improve readmission prediction. First, discriminative patterns that occur disproportionately between two classes: readmission and non-readmission, were generated based on hospital electronic health records. Second, the patterns were fed as features into a classification model for readmission prediction. Then, we have evaluated these discriminative pattern-based features in three datasets: diabetes, chronic kidney disease and all diseases. Experiments with each dataset showed that the features of chronic disease cohorts have fewer differences between the readmission and the non-readmission classes than the all-diseases cohort. Our proposed pattern-based model improved the prediction performance in terms of AUC (Area Under the receiver operating characteristic curve) by about 12% compared with the baseline models for the all-disease cohort, however, it showed little improvement for either diabetes or chronic kidney disease datasets.
Attaining the optimal scale size of production systems is an issue frequently found in the priority questions on management agendas of various types of organizations. Determining the most productive scale size (MPSS) ...
详细信息
Antenna arrays have a long history of more than 100 years and have evolved closely with the development of electronic and information technologies, playing an indispensable role in wireless communications and radar. W...
详细信息
Antenna arrays have a long history of more than 100 years and have evolved closely with the development of electronic and information technologies, playing an indispensable role in wireless communications and radar. With the rapid development of electronic and information technologies, the demand for all-time, all-domain, and full-space network services has exploded, and new communication requirements have been put forward on various space/air/ground platforms. To meet the ever increasing requirements of the future sixth generation (6G) wireless communications, such as high capacity, wide coverage, low latency, and strong robustness, it is promising to employ different types of antenna arrays (e.g., phased arrays, digital arrays, and reconfigurable intelligent surfaces, etc.) with various beamforming technologies (e.g., analog beamforming, digital beamforming, hybrid beamforming, and passive beamforming, etc.) in space/air/ground communication networks, bringing in advantages such as considerable antenna gains, multiplexing gains, and diversity gains. However, enabling antenna array for space/air/ground communication networks poses specific, distinctive and tricky challenges, which has aroused extensive research attention. This paper aims to overview the field of antenna array enabled space/air/ground communications and networking. The technical potentials and challenges of antenna array enabled space/air/ground communications and networking are presented first. Subsequently, the antenna array structures and designs are discussed. We then discuss various emerging technologies facilitated by antenna arrays to meet the new communication requirements of space/air/ground communication systems. Enabled by these emerging technologies, the distinct characteristics, challenges, and solutions for space communications, airborne communications, and ground communications are reviewed. Finally, we present promising directions for future research in antenna array enabled space/ai
We present a "calculator" for constructing a homogeneous approximation of nonlinear control systems, which is based on the algebraic approach developed by the authors in their previous papers. This approach ...
详细信息
In carrying out drilling projects at PT China Oilfields Services Limited ( COSL) Indo, especially Project 1 to Project 4, there were a mismatch between the initial plan of the project and the actualisation in the fiel...
In carrying out drilling projects at PT China Oilfields Services Limited ( COSL) Indo, especially Project 1 to Project 4, there were a mismatch between the initial plan of the project and the actualisation in the field because there were inhibiting factors in implementing the project. The purpose of the research was to look the factors that caused for the additional time in well drilling project at PT. COSL Indo, to build a relationship model of these factors, and formulate a strategy for the company to be able to overcome the occurrence of additional time in the future project. Data analysis was conducted by using a factor analysis method and the location of the research was carried out in just one company with 102 valid respondents. The results of the research indicate that there are four factors that influence the time gap, namely Project Management Activities, Risk Analysis and Procurement, Manage Stairs, and Project Planning Development. By knowing the causes of the delay, companies can find the best solutions for future learning and the impact on environmental, social and economic problems can also be anticipated so that sustainable development occurs in the context of petroleum management.
暂无评论