ID 原文 译文
57758 最后，基于真实数据集的实验验证了所提算法性能以及相应的缓存系统性能. Thereafter，real data sets to verify our prediction algorithm and investigate system performance are used.
57759 仿真结果表明，所提算法优于对比算法的缓存命中率性能. It is shown that the proposed scheme outperforms the comparison schemes in terms of hit ratio.
57760 机会移动社群网络易受到不良节点干扰而导致正常通信中断，现有研究方法普遍存在忽略不良行为差异性，为此，提出了基于狄利克雷分布的可信路由转发机制. The opportunistic mobile social network is vulnerable to the interruption of normal communica- tion due to the interference of bad nodes. Existing research methods generally have the problem of neglec- ting the difference of bad behavior. The article proposes a trusted routing and forwarding mechanism based on Dirichlet distribution.
57761 利用消息传递过程判断节点的可信度，提出应对干扰的路由转发机制. Firstly，the message passing process is used to judge the credibility of the node，and then the routing and forwarding mechanism for dealing with interference is proposed.
57762 实验结果表明，在受到不良节点干扰的条件下，该机制能够准确评估节点，同时在保持低传输成本的情况下，传输成功率比传统方法提高了 5% ～ 10% . Experi- ment shows that the mechanism can accurately evaluate nodes under the condition of poor node interfer- ence，and the transmission success rate is 5% ～ 10% higher than the traditional method while keeping the transmission cost low.
57763 针对传统日志模板挖掘时需要日志聚类数目作为先验信息的问题，提出了一种基于归一化特征判别的日志模板挖掘算法. A log template extraction algorithm based on normalized feature discrimination is proposed，ai- ming at the problem that the number of clusters needs to be provided as a priori information in traditional log template extraction.
57764 首先，对日志数据进行压缩，以提高后续处理效率; First，log data is initially compressed to reduce data redundancy.
57765 其次，进行日志聚类过程，使用归一化的日志统计特征判断聚类是否满足要求，若满足，则聚类成功; Then，a log clustering process is implemented，and the normalized feature is used to discriminate whether the cluste- ring result meets requirement:
57766 若不满足，则采用二分搜索的方式调整日志聚类的数目，重新进行聚类; if so，the clustering process is successfully completed; if not，the number of log clusters is adjusted by using binary search and redo clustering.
57767 最后，从聚类结果中提取日志模板，设计了一种衡量模板挖掘效果的评价指标. Finally，the log template is extrac- ted via clustering results. In addition，an evaluation metric that measures the effectiveness of template ex- traction is designed.