基于学习自动机与萤火虫算法的链路预测

便携交换网络由具有社区属性和移动规律的人组成,具有节点移动性、节点间间歇性连接、高延迟等特点,本文研究其网络行为预测中的链路预测问题,提出基于学习自动机和萤火虫算法的链路预测方法。采用学习自动机对节点进行自适应聚类,完成网络的社区划分；定义社区属性影响系数和移动行为影响系数,构建反映便携交换网络社区属性、节点移动性和节点间间歇性连接的相似性指标；将该指标与CN、RA、AA等指标融合,得到便携交换网络的相似性指标向量；借助差分整合移动平均自回归模型的时间序列分析能力,提取相似性指标向量序列的演化规律；采用萤火虫算法优化所构建的二分类器,预测节点对下一时刻的连接状态。在INFOCOM06和MIT两个真实数据集上的实验结果表明,与受限玻尔兹曼机、弱评估器等方法对比,本文方法具有更高的准确率和更好的稳定性。

Link Prediction Based on Learning Automaton and Firefly Algorithm

The pocket switch network (PSN) is composed of people with community attributes and mobility laws, which has lots of features such as node mobility, intermittent connection and high latency. This paper studies link prediction which is a part of network behavior prediction problem in PSN. A link prediction method is proposed, which is based on learning automaton and firefly algorithm. The learning automaton is employed to cluster nodes adaptively so as to complete the community division of the network. The node community attribute influence coefficient and mobile behavior influence coefficient are defined to construct the similarity index which reflects the community attribute, node mobility and intermittent connection between nodes. After fusing the index with CN, RA, AA, etc., a similarity vector of node pairs is achieved. Taking the advantages of analyzing time series of autoregressive integrated moving average model, the evolution law of the vector sequence is extracted. The binary classifier optimized by firefly algorithm is constructed to predict the connection of node pairs at the next moment. The experimental results on infocom06 and MIT datasets show that the proposed method has higher accuracy and better stability than the ones of RBM and week estimator.