In the process of protected protocol recognition,an improved AGglomerative NESting algorithm( IAGNES) with high adaptability is proposed,which is based on the AGglomerative NESting algorithm( AGNES),for the challenging issue of how to obtain single protocol data frames from multiprotocol data frames. It can improve accuracy and efficiency by similarity between bit-stream data frames and clusters,extract clusters in the process of clustering. Every cluster obtained contains similarity evaluation index which is helpful to evaluation. More importantly,IAGNES algorithm can automatically recognize the number of cluster. Experiments on the data set published by Lincoln Laboratory shows that the algorithm can cluster the protocol data frames with high accuracy.
在深入分析成像卫星任务规划问题模型要素的基础上,以有色Petri网为理论工具,提出一种基于CPN(Coloured Petri Net)的成像卫星任务规划模型。该模型分为顶层模型、控制模型、目标成像任务规划模型和图像传输任务规划模型,具有良好的模块化和通用化特点。通过设计典型仿真实例,考察所建模型在无任务冲突情况下和在有任务冲突情况下的任务规划能力,验证了所建模型的有效性。所建模型可以为成像卫星任务规划方案的制定提供理论依据。
现实量化交易应用中,传统的模糊数据挖掘算法往往需要针对给定的量化交易设定最小支持度阈值,然而,这些方法中存在的普遍问题是很难找到合适的最小支持度阈值,并且因为推导出的规则通常是常识而没有实际的商业意义。为了解决这个问题,提出了一种无需最小支持度阈值的模糊关联规则(fuzzy coherent rule,FCR)挖掘算法。首先将量化交易转换成模糊集,然后通过收集已经生成的模糊集生成候选模糊关联规则,最后计算出列联表并用其检查这些候选模糊关联规则是否满足四项判断准则。如果满足,则可以确定为模糊关联规则。在Foodmart数据集上的实验验证了所提算法的有效性,相比原始模糊关联规则(fuzzy association rules,FAR)挖掘算法,所提的FCR方法能够推导出更多的规则,并且能够在高置信度时推导出更多有用的规则。
