ID 原文 译文
3573 该技术能够显著降低访问控制策略生成的时间成本，为访问控制的实施提供有效支持。 This technology could significantly reduce the time cost of access control policy generation and pro-vide effective support for the implementation of access control.
3574 将策略生成问题分解为访问控制语句识别和访问控制属性挖掘两项核心任务，分别设计了 BiGRU-CNN-Attention 和 AM-BiLSTM-CRF 这 2 个神经网络模型来实现访问控制策略语句识别和访问控制属性挖掘，从而生成可读、可执行的访问控制策略。 The policy generation problem was decomposed into twocore tasks, identification of access control policy sentence and access control attribute mining. Neural network modelssuch as BiGRU-CNN-Attention and AM-BiLSTM-CRF were designed respectively to realize identification of accesscontrol policy sentence and access control attribute mining, so as to generate readable and executable access control poli-cies.
3575 实验结果表明，与基准方法相比，所提方法具有更好的性能。 Experimental results show that the proposed method has better performance than the benchmark method.
3576 特别是在访问控制策略语句识别任务中平均 F1-score 指标能够达到 0.941，比当前的 state-of-the-art 方法性能提高了 4.1%。 In particu-lar, the average F1-score index can reach 0.941 in the identification task of access control policy sentence, which is 4.1% better than the current state-of-the-art method.
3577 基于改进蚁群优化算法与子图演化，提出了一种新型非监督社交网络链路预测（SE-ACO）方法。 Based on improved ant colony algorithm and subgraph evolution fusion, a new unsupervised social networklink prediction method (SE-ACO) was proposed.
3578 该方法首先在社交网络图中确定特殊子图； First, the special subgraph was determined in the social network graph.
3579 然后研究子图演化以预测图中的新链接，并用蚁群优化算法定位特殊子图； Then the evolution of the subgraph was studied to predict the new links in the graph, and the special subgraph was lo-cated by the ant colony method.
3580 最后针对所提方法使用不同网络拓扑环境与数据集进行检验。 Finally, using different network topology environments and data sets to test the proposedmethod.
3581 结果表明，与其他无监督社交网络预测算法相比，所提SE-ACO 方法在多数数据集上的评估结果较好，且运行时间较短，这表明图形结构在链路预测算法中起重要作用。 Compared with other unsupervised social network prediction algorithms, the proposed SE-ACO method has thebest evaluation results, shorter running time and the best effect on most data sets, which indicates that graph structureplays an important role in link prediction algorithm.
3582 针对传统集中式贸易系统中管理者滥用权力问题，提出面向区块链贸易系统的无管理者安全模型，同时解决去管理者的贸易系统由于管理员缺失导致的背书安全、贸易不及时、验证低效、动态低效等问题。 In view of the abuse of power by managers in the traditional centralized trading system, a security model without managers for the blockchain-based trading system was proposed, which also solved the problems of unsafe en-dorsement, untimely trading, low auditing efficiency and dynamic inefficiency caused by the elimination of managers.