Search

The hybrid neural network model proposed in this paper consists of two main parts: extracting local features of text vectors by convolutional neural network, extracting global features related to text context by BiLSTM, and fusing the features extracted by the two complementary models. In this paper, the pre-processed sentences are put into the hybrid neural network for training. The trained hybrid neural network can automatically classify the sentences. When testing the algorithm proposed in this paper, the training corpus is Word2vec. The test results show that the accuracy rate of text categorization reaches 94.2%, and the number of iterations is 10. The results show that the proposed algorithm has high accuracy and good robustness when the sample size is seriously unbalanced.

When combining very different images which often contain complex objects and backgrounds, producing consistent compositions is a challenging problem requiring seamless image editing. In this paper, we propose a general approach, called object-aware image editing, to obtain consistency in structure, color, and texture in a unified way. Our approach improves upon previous gradient-domain composition in three ways. Firstly, we introduce an iterative optimization algorithm to minimize mismatches on the boundaries when the target region contains multiple objects of interest. Secondly, we propose a mixed-domain consistency metric for measuring gradients and colors, and formulate composition as a unified minimization problem that can be solved with a sparse linear system. In particular, we encode texture consistency using a patch-based approach without searching and matching. Thirdly, we adopt an object-aware approach to separately manipulate the guidance gradient fields for objects of interest and backgrounds of interest, which facilitates a variety of seamless image editing applications. Our unified method outperforms previous state-of-the-art methods in preserving global texture consistency in addition to local structure continuity.

In regulated freshwater ecosystems, the deposition process of organic carbon (OC) in sediment could be affected by all the changes of physical, chemical, and biological factors. To assess different hydrodynamic conditions on the deposition of OC in sediment, the combination of elemental and stable isotope analyses was carried out in two reservoirs: the Huairou Reservoir (HR), which undergoes seasonal water level fluctuation (WLF) like natural lakes, and the Shisanling Reservoir (SR), which undergoes daily WLF for the operation of pumped storage power station. In general, OC concentration and its variation were comparable in the two reservoirs, whereas the OC distribution showed completely different features. The OC burial rate in SR was nearly two times more than in the HR, which reflect the impact of WLFs on the erosion of soil. For the relative contribution of phytoplankton-derived OC (P-OC), macrophyte-derived OC (M-OC), and terrestrial OC (T-OC) in sediment, the T-OC and M-OC in the SR were both higher than those of the HR, whereas the P-OC was markedly lower. Furthermore, the classification of nearshore and offshore sediments suggested that hydrodynamic conditions affected the T-OC import by physical processes and the in situ contributions of P-OC and M-OC with a spatial difference.

Protein phosphatase 2A (PP-2A) has been implicated to be crucial in neural development and the normal function of nervous system. However, little is known about its role in neuritogenesis. In this study, we reported that inhibition of PP-2A strongly suppresses the outgrowth of cell processes only during the initiation stage, while activation of PP-2A promotes extensive outgrowth of long neurites in Neura2A cells and long single axon or multiple axons in hippocampal neurons. Our results indicated that PP-2A may be an important positive regulator in neurite outgrowth, and upregulation PP-2A could be a possible target for the therapy of axonopathy in neural diseases.