ID 原文 译文
16985 常用的异质信息网络有知识图谱和具有简单模式层的异质信息网络，它们的表示学习通常遵循不同的方法。 Commonly used heterogeneous information networks include knowledge graphs and heterogeneousinformation networks with simple schemas. Their representation learning follows usually different methods.
16986 该文总结了知识图谱和具有简单模式层的异质信息网络之间的异同，提出了一个通用的异质信息网络表示学习框架。 The similarities and differences between knowledge graphs and heterogeneous information networks with simple schemas are summarized, and a general heterogeneous information network representation learning framework is proposed.
16987 该文提出的框架可以分为3个部分：基础向量模型，基于图注意力网络的传播模型以及任务模型。 The proposed framework can be divided into three parts: the basic vector model, the graphattention network based propagation model, and the task model.
16988 基础向量模型用于学习基础的网络向量；传播模型通过堆叠注意力层学习网络的高阶邻居特征； The basic vector model is used to learn basicnetwork vector; The propagation model learns the high-order neighbor features of the network by stackingattention layers.
16989 可更换的任务模型适用于不同的应用场景。 The replaceable task module is suitable for different application scenarios.
16990 与基准模型相比，该文所提框架在知识图谱的链接预测任务和异质信息网络的节点分类任务中都取得了相对不错的效果。 Compared with thebenchmark model, the proposed framework achieves relatively good results in the link prediction task of theknowledge graph and the node classification task of the heterogeneous information network.
16991 针对磁共振图像(MRI)进行脑胶质瘤检测及病灶分割对临床治疗方案的选择和手术实施过程的引导都有着重要的价值。 The glioma detection and focus segmentation in Magnetic Resonance Imaging (MRI) has importantvalue for the therapeutic schedule selection and the surgical operations.
16992 为了提高脑胶质瘤的检测效率和分割准确率，该文提出了一种两阶段计算方法。 In order to improve the detectionefficiency and segmentation accuracy for glioma, this paper proposes a two-stage calculating method.
16993 首先，设计了一个轻量级的卷积神经网络，并通过该网络完成MR图像中肿瘤的快速检测及大致定位； First, alight convolutional neural network is designed to implement rapidly detection and localization for the glioma inMR images.
16994 接着，通过集成学习过程对肿瘤周围水肿、肿瘤非增强区、肿瘤增强区和正常脑组织等4种不同区域进行分类与彼此边界的精细分割。 Then, the peritumoral edema, non-enhancing tumor, enhancing tumor, and normal are classifiedand segmented from each other through an Ensemble Learning (EL) process.