CAJ | 학술논문

针对已有测度无法准确评价网络抗毁度和节点重要性的问题，采用了一种基于网络邻接矩阵谱特征的抗毁测度。抗毁测度的值越大，网络中替代路径就越多，网络的抗毁能力越强。在此基础上，用克隆选择算法对网络的抗毁性进行优化。实验结果表明新抗毁测度是有效的，优化算法也能大大提高网络的抗毁性。
침대이유측도무법준학평개망락항훼도화절점중요성적문제，채용료일충기우망락린접구진보특정적항훼측도。항훼측도적치월대，망락중체대로경취월다，망락적항훼능력월강。재차기출상，용극륭선택산법대망락적항훼성진행우화。실험결과표명신항훼측도시유효적，우화산법야능대대제고망락적항훼성。
The network invulnerability and node importance coued not be evaluated effectively. So a new measure by the current measures is proposed based on the spectrum feature of network adjacent matrix. It shows that the larger the invulnerable measure is, the more alternative routes present, andthe better the network invulnerability is. Then a clonal selection algorithm is used to improve the net- work invulnerability. The experiment also shows the new measure is effective and the invulnerability is improved greatly.