ID 原文 译文
57878 设计了一个捕捉区块内相邻会话关联关系的区块级 ＲNN，进一步描述用户的中期兴趣; Next，to further describe medi- um-term interests，a block-level ＲNN is added to capture correlations across sessions in a block.
57879 使用一个用户级 ＲNN 追踪长期兴趣的演化; Then，a user-level ＲNN is devised to track evolution of long-term interests.
57880 引入带有不同交互机制的融合层，以有效融合不同层次的兴趣信息. Finally，the article designs fusion lay- ers with different interaction mechanisms to effectively integrate cross-level interest information.
57881 在 3 个真实数据集上进行实验，结果表明，该方法与先进的推荐方法相比，Ｒecall@ 10 提升了 18.35% . Simula- tions on three real-world datasets show that PHＲM outperforms the state-of-the-art recommendation meth- ods，with Ｒecall@ 10 increasing by 18. 35% .
57882 针对段落式图像描述生成研究中提升描述语句之间的连贯性问题，提出了一种基于全卷积结构的图像段落描述算法. How to improve the coherence among descriptive sentences for the paragraph image captioning is paid attention currently. A fully convolutional neural architecture for paragraph image captioning was proposed.
57883 采用基于卷积网络的区域检测器获取图像表示，结合段落在语言学角度的层次性，构建一种层次性的深度卷积解码器对图像表示解码，自动生成段落式文本描述. An image representation is first obtained using a region detector based on a convolutional net- work. Then a hierarchical deep convolutional decoder is constructed to translate the image representation， automatically generating a paragraph text description.
57884 同时将门控机制嵌入卷积解码器网络中，以提升模型的记忆能力. In addition，the gating mechanism is embedded in the convolutional decoder network to improve memory capacity of the model.
57885 实验结果表明，相比于基于循环神经网络等传统段落图像的描述方法，新算法能够为图像生成更为连贯的段落式文本描述，在评测指标上取得较好的结果. Experiments demonstrate that compared with those traditional methods based on recurrent neural networks，the proposed algorithm can generate more coherent paragraph text descriptions for images，achieving better results on evaluation metrics.
57886 锚节点的布设方案决定了无线传感器网络的定位极限. The anchor node layout scheme determines the performance of localization in the wireless sen- sor network.
57887 针对现有布设算法存在计算开销大和优化策略具有局限性等问题，提出了基于天牛须搜索的锚节点布设优化算法. In order to solve the problems of high computational cost and limited optimization strategy of existing layout algorithms，an anchor node layout optimization algorithm based on the beetle antennae search was proposed.