We present PARABANK, a large-scale English paraphrase dataset that surpasses prior work in both quantity and quality. Following the approach of PARANMT (Wieting Gimpel, 2018), we train Czech-English neural machine translation (NMT) system to generate novel paraphrases reference sentences. By adding lexical constraints NMT decoding procedure, however, are able produce multiple high-quality sentential per source sentence, yielding an resource with more than 4 billion generated tokens...

Cardiovascular Disease (CVD) is a highly significant contributor to loss of quality and quantity life all over the world. Early detection risk prediction very important for patients' treatment doctors' diagnose. This paper focus on establishing more accurate practical system based data mining techniques provide auxiliary medical service. In order be practically used collecting analyzing in healthcare industries, consists four parts: interface, preparation, feature selection classification....

<title>Abstract</title> <bold>Chronic postsurgical pain (CPSP) is prevalent after distal lower limb fracture. To identify high-risk patients early and provide individualized treatment, we aimed to develop validate a clinical risk model that can predict CPSP in followed for at least 3 months surgery. In this retrospective cohort study, identified who required open reduction internal fixation tibial, fibular, or ankle fractures up We focused on variables accessible clinicians during the...

Road traffic forecasting plays a critical role in smart city initiatives and has experienced significant advancements thanks to the power of deep learning capturing non-linear patterns data. However, promising results achieved on current public datasets may not be applicable practical scenarios due limitations within these datasets. First, limited sizes them reflect real-world scale networks. Second, temporal coverage is typically short, posing hurdles studying long-term acquiring sufficient...

Aspect-term sentiment analysis (ATSA) is a long-standing challenge in natural language process. It requires fine-grained semantical reasoning about target entity appeared the text. As manual annotation over aspects laborious and time-consuming, amount of labeled data limited for supervised learning. This paper proposes semi-supervised method ATSA problem by using Variational Autoencoder based on Transformer. The model learns latent distribution via variational inference. By disentangling...

Meningioma is the second most commonly encountered tumor type in brain. There are three grades of meningioma by standards World Health Organization. Preoperative grade prediction extraordinarily important for clinical treatment planning and prognosis evaluation. In this paper, we present a new deep learning model assisting automatic to reduce recurrence meningioma. Our based on an improved LeNet-5 convolutional neural network (CNN) does not require extraction diseased tissue, which can...

Sensors are the key to environmental monitoring, which impart benefits smart cities in many aspects, such as providing real-time air quality information assist human decision-making. However, it is impractical deploy massive sensors due expensive costs, resulting sparse data collection. Therefore, how get fine-grained measurement has long been a pressing issue. In this paper, we aim infer values at non-sensor locations based on observations from available (termed spatiotemporal inference),...

A majority of metro projects have been constructed to reduce urban traffic congestion and improve the convenience public transportation, but these also produced a significant amount engineering slag mud. The shield construction method could efficiency safety; this technique has frequently used in tunnel excavation projects. However, spoil during is challenging deal with. In literature, though there an increasing number studies on technologies utilizing spoil, on-site utilization still...

We study the task of spatio-temporal extrapolation that generates data at target locations from surrounding contexts in a graph. This is crucial as sensors collect are sparsely deployed, resulting lack fine-grained information due to high deployment and maintenance costs. Existing methods either use learning-based models like Neural Networks or statistical approaches Gaussian Processes for this task. However, former lacks uncertainty estimates latter fails capture complex spatial temporal...

In machine translation evaluation, a good candidate can be regarded as paraphrase of the reference. We notice that some words are always copied during paraphrasing, which we call copy knowledge. Considering stability such knowledge, should contain all these appeared in reference sentence. Therefore, this participation WMT’2018 metrics shared task introduce simple statistical method for knowledge extraction, and incorporate it into Meteor metric, resulting new metric Meteor++. Our experiments...

Abstract Background Malonylation is a recently discovered post-translational modification that associated with variety of diseases such as Type 2 Diabetes Mellitus and different types cancers. Compared experimental identification malonylation sites, computational method time-effective process comparatively low costs. Results In this study, we proposed novel model called Mal-Prec (Malonylation Prediction) for site prediction through the combination Principal Component Analysis Support Vector...

The problem of community detection has received great attention in recent years. Many methods have been proposed to discover communities networks. In this paper, we propose a Gaussian stochastic blockmodel that uses distributions fit weight edges networks for non-overlapping detection. maximum likelihood estimation model the same objective function as general label propagation with node preference. preference specific vertex turns out be value proportional intra-community eigenvector...

Power quality issues have become one of the most important issue for researchers to concern. In this paper, simulation and experiment algorithm circuit design Unified Quality Conditioner (UPQC) are provided. Control topology UPQC which includes active power filter (APF) dynamic voltage restorer (DVR) introduced. Stability condition unit is deduced proved by Routh stability criterion. Simulation APF DVR carried out in PSCAD show proposed control strategy. Experiments such as current tracking,...

Recently, doc2vec has achieved excellent results in different tasks. In this paper, we present a context aware variant of doc2vec. We introduce novel weight estimating mechanism that generates weights for each word occurrence according to its contribution the context, using deep neural networks. Our model can achieve similar compared initialized byWikipedia trained vectors, while being much more efficient and free from heavy external corpus. Analysis shows they are kind enhanced IDF capture...

The primary goal of skeletal motion prediction is to generate future by observing a sequence 3D skeletons. A key challenge in the fact that can often be performed several different ways, with each consisting its own configuration poses and their spatio-temporal dependencies, as result, predicted converge motionless or non-human like motions long-term prediction. This leads us define hierarchical recurrent network model explicitly characterizes these internal configurations local global...