ID 原文 译文
56178 链路预测任务是根据已知的网络结构和节点属性等信息来预测网络中产生新链路的可能性. Link prediction is the task of forecasting the possibility of generating new links in a network basedon network structure and node attributes.
56179 它是网络科学中的一个基础性问题,具有重要的理论研究和实际应用价值. It is a basic problem in network science and valuable both in theoryand practice.
56180 近年来,网络表示学习领域的学者利用深度学习提取网络复杂特征,大幅度提高了链路预测效果. In recent years, deep learning methods are widely used for network representation to extractcomplex network features, which greatly improves the results of link prediction.
56181 实际网络中节点具有局部聚类现象,然而,当前的网络表示学习侧重于提取网络全局特征,忽略了局部信息特征. The nodes in real-world networkshave the phenomenon of local clustering; however, the current network representation learning methods focuson extracting the global features of the network, ignoring the local information features.
56182 针对这个问题,我们提出了能够学习网络中节点在不同社区中局部特征表示的模型network-splitter. In order to solvethis problem, we propose a network-splitter model, which can learn the local feature representation of nodes indifferent communities.
56183 该模型利用重叠社区思想,在每个社区中创建节点的一个角色副本,并学习该角色副本的特征表示. The model uses the idea of overlapping community to create a role copy of a node ineach community and learns the feature representation of the role copy.
56184 最后将节点在不同社区中对应的角色副本信息通过神经网络综合,得到的综合向量包含网络全局特征和节点局部特征,并可应用到链路预测任务中. Finally, the role copy information of thenode in different communities is synthesized through the neural network. Using this method, both the globalfeature of the network and the local features of the nodes are summarized into the network representation, andthen are applied to the link prediction.
56185 本文的实验结果表明, network-splitter模型与最新的网络学习表示方法相比具有很强的竞争力. The experimental results show that the network-splitter model has strongcompetitiveness compared with the latest network representation learning methods
56186 在复杂网络研究领域,社区隐藏算法旨在破坏社区发现算法的有效性,从而隐藏用户之间的关系,达到保护用户隐私的目的. Community deception algorithms aim to disable the community detection algorithms and thus hiderelationships among users in social networks.
56187 现有的社区隐藏算法寄希望于已有用户改变自己的社交关系实现该目标,这对用户影响较大,操作空间比较小.不同于此类算法,本文所提的算法从增加节点及其相应边的角度出发,即伪造用户及其关系,最大程度降低了对用户的影响,将社区隐藏问题转换为网络增长问题. Previous studies on community deception focus on modifying existinggraph structures, which is infeasible in most real-world applications. In contrast, it is more practical to injectnodes and edges into existing graphs.