ID 原文 译文
44376 社区发现能有效挖掘网络的特性以及隐藏的信息。 Community detection can effectively mine the characteristics of the network as well as the hidden information.
44377 局部扩展是社区发现常用的一种方法，该方法大体上可以分为种子的选择和局部扩展两部分。 Local expansion is a commonly used method of community detection, and it can be divided into two steps: the selection of seeds and the local expansion.
44378 因此，为了分析现有方法的优劣以及适用场合，对种子的选择、局部扩展以及评价指标等方法进行概括、比较和分析，总结了基于局部扩展的社区发现的应用以及研究难点。 Therefore, in order to analyze the advantages and disadvantages of the existing methods and their application, these methods about the selection of seeds, local expansion and evaluation were summarized, compared and analyzed. Then, the application and the research difficulties of community detection based on local extension were summarized.
44379 最后，对基于局部扩展的社区发现的研究方向进行了展望。 Finally, the research directions of community detection based on local expansion were given.
44380 为提高弱监督语义分割算法精度，提出一种融合多尺度特征的分割及优化算法。 In order to improve the accuracy of weakly-supervised semantic segmentation method, a segmentation and optimization algorithm that combines multi-scale feature was proposed.
44381 首先，基于迁移学习算法构建多尺度特征模型，类别预测时引入新分类器，减少因预测目标类信息错误导致分割失败的情况； The new algorithm firstly constructs a multi-scale feature model based on transfer learning algorithm. In addition, a new classifier was introduced for category prediction to reduce the failure of segmentation due to the prediction of target class information errors.
44382 其次，将多尺度模型与原迁移学习模型进行加权集成，增强模型泛化性能； Then the designed multi-scale model was fused with the original transfer learning model by different weights to enhance the generalization performance of the model.
44383 最后，结合预测类可信度调整分割图中相应类像素的可信度，规避假正例分割区域。 Finally, the predictions class credibility was added to adjust the credibility of the corresponding class of pixels in the segmentation map, avoiding false positive segmentation regions.
44384 在 VOC 2012 验证集上的平均交并比为 58.8%，测试集上的平均交并比为57.5%，同比原迁移学习模型分别提升 12.9%和 12.3%，也优于其他以类标作为监督信息的语义分割算法。 The proposed algorithm was tested on the challenging VOC 2012 dataset, the mean intersection-over-union is 58.8% on validation dataset and 57.5% on test dataset. It outperforms the original transfer-learning algorithm by 12.9% and 12.3%. And it performs favorably against other segmentation methods using weakly-supervised information based on category labels as well.
44385 为提高定位效率，提出了一种基于多分布密度位置指纹、精度渐进的室内定位算法。 To improve the prediction speed in indoor localization, a novel algorithm based on fingerprint with varied scales was proposed.