Multi-temporal change detection (CD) plays a crucial role in the remote sensing application field. In recent years, supervised deep learning methods have shown excellent performance detecting changes very-high-resolution (VHR) images. However, these require large number of labeled samples for training, making process time-consuming and labor-intensive. Unsupervised approaches are more attractive practical applications since they can produce CD map without relying on any ground reference or...

Abstract Although drug combinations in cancer treatment appear to be a promising therapeutic strategy with respect monotherapy, it is arduous discover new synergistic due the combinatorial explosion. Deep learning technology holds immense promise for better prediction of vitro certain cell lines. In methods applying such technology, omics data are widely adopted construct line features. However, biological network rarely considered yet, which worthy in-depth study. this study, we propose...

Identification of drug-target interactions (DTIs) is an important step in drug discovery and repositioning. In recent years, graph-based methods have attracted great attention show advantages on predicting potential DTIs. However, these face the problem that known DTIs are very limited expensive to obtain, which decreases generalization ability methods. Self-supervised contrastive learning independent labeled DTIs, can mitigate impact problem. Therefore, we propose a framework SHGCL-DTI for...

The commit messages in source code repositories are valuable but not easy to be generated manually time for tracking issues, reporting bugs, and understanding codes. Recently published works indicated that the deep neural machine translation approaches have drawn considerable attentions on automatic generation of messages. However, they could deal with out-of-vocabulary (OOV) words, which essential context-specific identifiers such as class names method diffs. In this paper, we propose...

The electrification of shared mobility has become popular across the globe. Many cities have their new e-mobility systems deployed, with continuously expanding coverage from central areas to city edges. A key challenge in operation these is fleet rebalancing, i.e., how EVs should be repositioned better satisfy future demand. This particularly challenging context systems, because i) range limited while charging time typically long, which constrain viable rebalancing operations; and ii) EV...

Abstract Motivation Drug combination therapy shows significant advantages over monotherapy in cancer treatment. Since the combinational space is difficult to be traversed experimentally, identifying novel synergistic drug combinations based on computational methods has become a powerful tool for pre-screening. Among them, deep learning have far outperformed other methods. However, most learning-based are unstable and will give inconsistent predictions even by simply changing input order of...

Intelligent classrooms have demonstrated significant promise in enhancing learning efficiency as a result of the quick development big data and artificial intelligence technologies. This study proposes text semantic matching model (OM) that combines deep K-means clustering algorithm, aiming to optimize vocabulary. Importantly, it delves into biomechanical aspects by considering how physical physiological processes interact with language acquisition. By mimicking mechanism biological neural...

In change detection tasks, seasonal variations in spectral characteristics and surface cover can negatively impact performance when comparing image pairs from different seasons. Many existing methods do not specifically address the degradation caused by errors. To tackle this issue, Dual-Branch Seasonal Error Elimination Change Detection Framework using Target Image Feature Fusion Generator (DBSEE-CDF) is introduced. Specifically, approach utilizes (TIFFG), which incorporates spatial channel...

The future of urban mobility is expected to be shared and electric. It not only a more sustainable paradigm that can reduce emissions, but also bring societal benefits by offering affordable on-demand option the general public. Many car sharing service providers as well automobile manufacturers are entering competition expanding both their EV fleets renting/returning station networks, aiming seize share market zero emissions level. During fast expansion, one determinant for success ability...

Electric Vehicle (EV) sharing systems have recently experienced unprecedented growth across the world. One of key challenges in their operation is vehicle rebalancing, i.e., repositioning EVs stations to better satisfy future user demand. This particularly challenging shared EV context, because i) range limited while charging time substantial, which constrains rebalancing options; and ii) as a new mobility trend, most current are still continuously expanding station networks, targets for can...

Pansharpening aims at fusing the rich spectral information of multispectral (MS) images and spatial details panchromatic (PAN) to generate a fused image with both high resolutions. In general, existing pansharpening methods suffer from problems distortion lack detail information, which might prevent accuracy computation for ground object identification. To alleviate these problems, we propose Hybrid Attention mechanism-based Residual Neural Network (HARNN). proposed network, develop an...

Multistep flow prediction is an essential task for the car-sharing systems. An accurate model can help system operators to pre-allocate cars meet demand of users. However, this challenging due complex spatial and temporal relations among stations. Existing works only considered (e.g. using LSTM) or CNN) independently. In paper, we propose attention multi-graph convolutional sequence-to-sequence (AMGC-Seq2Seq), which a novel deep learning multistep prediction. The proposed uses...

In different types of feature selection algorithms, clustering is an emerging subset generation paradigm. this paper, a Minimum spanning tree based Feature Clustering (MFC) algorithm proposed. the algorithm, information-theoretic measure, i.e., Variation information, utilized as redundancy and relevance metric. At phase, sum pair wise minimized. Then, representative selected from each cluster, where between features target label maximized. The supervised since it designed for various...

The following topics are dealt with: geophysical image processing; remote sensing; classification; time series; learning (artificial intelligence); feature extraction; radar imaging; synthetic aperture radar; resolution; convolutional neural nets.

The structure of the data relationship is complex, and this complex exists among multiple attributes, entity relations, entities at-tributes, even between database. In paper, we improve a visualization method. On basis original visual structure, dynamically changed according to user's needs. Besides, relationships can change dynamically, depending on data. At same time, use improving method query database instead learning SQL. proposed in paper not limited by set, any set Finally, plan...

With the rapid growth of multimodal data in social medias and huge requirement short but abundant information. Multimodal summarization has drawn much attention both industry academia. It usually obtains textual summary from multiple sources by computer vision or nature language processing technologies. However, there are also two challenges modeling such task: 1) The feature representation is limited non-alignment among data; 2) Massive parallel required during training, which...

Plagiarism is a matter of great concern in academic. Especially, colleges and universities which involve homework, exams, credits degrees, are the hardest hit area for plagiarism. Although it common sense about plagiarism teachers students that forbidden, blurred boundary line always there when happens real cases. In computer science, problem more severe as definition code ambiguous difficult than text The assignments projects science include not only documents, but also codes, contain...

Over the past few years, deep learning algorithms have held immense promise for better multi-spectral (MS) optical remote sensing image (RSI) analysis. Most of proposed models, based on convolutional neural network (CNN) and fully (FCN), been applied successfully computer vision images (CVIs). However, there is still a lack exploration spectra correlation in MS RSIs. In this study, with spectrum separable module (DSSM) semantic segmentation, which enables utilization characteristics The...