ID	原文	译文
4994	为了使 Kubernetes 集群对部署在其上的应用资源使用量能“提前”响应，并根据预测值为应用及时、准确、动态地调度和分配资源，提出了一种基于三次指数平滑法和时间卷积网络的云资源预测模型，根据历史数据预测未来的资源需求。	In order to allow the Kubernetes cluster to respond “in advance” to the resource usage of the applications deployed on it, and then to schedule and allocate resources in a timely, accurate and dynamic mannerbased on the predicted value, a cloud resource prediction model based on triple exponential smoothing method and tem-poral convolutional network was proposed, based on historical data to predict future demand for resources.
4995	为了找到参数的最优组合，使用 TPOT 调参思想对参数进行优化。	To find theoptimal combination of parameters, the parameters were optimized using TPOT thought.
4996	对 Google 数据集 CPU 和内存的预测实验表明，所提模型与其他模型相比具有更好的预测性能。	Experiments on the CPU andmemory of the Google dataset show that the model has better prediction performance than other models.
4997	孤立森林算法是基于隔离机制的异常检测算法，存在与轴平行的局部异常点无法检测、对高维数据异常点缺乏敏感性和稳定性等问题。	The isolation-based anomaly detector, isolation forest has two weaknesses, its inability to detect anomalies thatwere masked by axis-parallel clusters, and anomalies in high-dimensional data.
4998	针对这些问题，提出了基于随机超平面的隔离机制和多粒度扫描机制，随机超平面使用多个维度的线性组合简化数据模型的隔离边界，利用随机线性分类器的隔离边界能够检测更复杂的数据模式。	An isolation mechanism based on random hyperplane and a multi-grained scanning was proposed to overcome these weaknesses. The random hyperplane generatedby a linear combination of multiple dimensions was used to simplify the isolation boundary of the data model which was a random linear classifier that can detect more complex data patterns, so that the isolation mechanism was more consis-tent with data distribution characteristics.
4999	同时，多粒度扫描机制利用滑动窗口的方式进行维度子采样，每一个维度子集均训练一个森林，多个森林集成投票决策，构造层次化集成学习异常检测模型。	The multi-grained scanning was used to perform dimensional sub-sampling which trained multiple forests to generate a hierarchical ensemble anomaly detection model.
5000	实验表明，改进的孤立森林算法对复杂异常数据模式有更好的稳健性，层次化集成学习模型提高了高维数据中异常检测的准确性和稳定性。	Experiments show that the improved isolation forest has better robustness to different data patterns and improves the efficiency of anomaly points inhigh-dimensional data.
5001	为解决广域信息管理（SWIM）服务提供者由于自身故障或受到恶意攻击，造成 SWIM 服务中断、服务时延增加或服务质量下降的问题，提出了一种基于态势感知的 SWIM 服务权限主动移交模型，利用随机森林算法判别 SWIM 服务提供者安全态势，依据安全态势主动移交 SWIM 服务权限，降低突发事件对 SWIM 服务的影响。	To solve the problem that in the system wide information management (SWIM) network, the SWIM serviceprovider suffers from SWIM service interruption, service delay increase or service quality degradation due to maliciousattack or self-failure. Therefore, a proactive migration model of SWIM service was proposed on the basis of situationawareness, which used the random forest algorithm to timely judge the SWIM service provider security situation. SWIMservice authority was actively migrated according to security situation, and the emergencies impact on SWIM serviceswere reduces.
5002	实验证明，所提模型能够在突发事件发生的情况下保证服务的连续性，与未部署服务移交模型的 SWIM 网络相比，具有更高的可靠性和稳定性。	Experimental results show that the proposed model can guarantee services continuity in an emergencyevent, which has higher reliability and stability than SWIM network in which the service migration model is not dep-loyed.
5003	为了应对网络切片中共享物理资源的虚拟网络功能（VNF）间因资源竞争带来的性能下降问题，提出了一种基于性能感知的网络切片部署方法。	In order to deal with the performance degradation caused by resource contention due to the sharing of physicalresources between VNF in the network slicing, a network slicing deployment method based on performance-awarenesswas proposed.