A New BRB Model for Cloud Security-State Prediction Based on the Large-Scale Monitoring Data

作者：Wei, Hang; Hu, Guan-Yu; Han, Xiaoxia; Qiao, Peili^*; Zhou, Zhiguo; Feng, Zhi-Chao; Yin, Xiao-Jing

来源：IEEE Access, 2018, 6: 11907-11920.

DOI：10.1109/ACCESS.2017.2779599

• Belief rule base (BRB)
• cloud computing
• multi-attributes integration
• security-state prediction

Considering the reliability of the cloud computing system, this paper aims to predict the security state with multiple large-scale attributes in cloud computing system. A double-layer method for predicting the security state of cloud computing system based on the belief rule-base model is proposed, where the evidential reasoning (ER) algorithm is employed to fuse the multiple system indicators of actual cloud system and make a reasonable assessment to describe the cloud security state. This method can utilize quantitative and qualitative information simultaneously. By using the ER algorithm to integrate multiple indicators whose attributes contain much uncertain information, the security state of the cloud computing system can be predicted accurately. Moreover, due to the initial parameters of the proposed models are given by experts that might cause imprecise results, the constraint CMA-ES algorithm is employed as the optimization tool to obtain the optimal parameters. A practical study about the cloud security-state prediction is verified to indicate the potential applications about the proposed prediction model in a cloud computing platform.

• 单位
  哈尔滨学院; 海南师范大学; 长春工业大学