为探究全服务航空公司航线网络演化影响因素及演化模型,以全球排名前10的全服务航空公司为研究对象,基于链路预测思想,对5种内生属性相似性指标和3种外生属性相似性指标及耦合相似性指标分别进行曲线下面积(Area Under Curve,AUC)计算,...为探究全服务航空公司航线网络演化影响因素及演化模型,以全球排名前10的全服务航空公司为研究对象,基于链路预测思想,对5种内生属性相似性指标和3种外生属性相似性指标及耦合相似性指标分别进行曲线下面积(Area Under Curve,AUC)计算,选取AUC值最高的相似性指标进行航线预测及验证,并构建基于链路预测的全服务航空公司航线网络改进BBV动态加权演化模型.研究结果表明:基于优先连接(Preferential Attachment,PA)指标和机场航班量指标的耦合相似性指标的AUC值最高,是影响全服务航空公司航线网络演化的关键指标;在基于该耦合相似性指标的链路预测算法中,平均航线命中率达45.79%,AUC值达95%以上;相比于传统BBV模型,改进BBV动态加权演化模型的航线数拟合准确率平均提升了4.43%,能够较为科学地拟合不同全服务航空公司航线网络的扩张与收缩演变过程.展开更多
Community structure is an important characteristic in real complex network.It is a network consists ofgroups of nodes within which links are dense but among which links are sparse.In this paper, the evolving network i...Community structure is an important characteristic in real complex network.It is a network consists ofgroups of nodes within which links are dense but among which links are sparse.In this paper, the evolving network includenode, link and community growth and we apply the community size preferential attachment and strength preferentialattachment to a growing weighted network model and utilize weight assigning mechanism from BBV model.Theresulting network reflects the intrinsic community structure with generalized power-law distributions of nodes'degreesand strengths.展开更多
文摘为探究全服务航空公司航线网络演化影响因素及演化模型,以全球排名前10的全服务航空公司为研究对象,基于链路预测思想,对5种内生属性相似性指标和3种外生属性相似性指标及耦合相似性指标分别进行曲线下面积(Area Under Curve,AUC)计算,选取AUC值最高的相似性指标进行航线预测及验证,并构建基于链路预测的全服务航空公司航线网络改进BBV动态加权演化模型.研究结果表明:基于优先连接(Preferential Attachment,PA)指标和机场航班量指标的耦合相似性指标的AUC值最高,是影响全服务航空公司航线网络演化的关键指标;在基于该耦合相似性指标的链路预测算法中,平均航线命中率达45.79%,AUC值达95%以上;相比于传统BBV模型,改进BBV动态加权演化模型的航线数拟合准确率平均提升了4.43%,能够较为科学地拟合不同全服务航空公司航线网络的扩张与收缩演变过程.
基金Supported by the National Nature Science Foundation of China under Grant No.10832006PuJiang Project of Shanghai under Grant No.09PJ1405000+1 种基金Key Disciplines of Shanghai Municipality (S30104)Research Grant of Shanghai University under Grant No.SHUCX092014
文摘Community structure is an important characteristic in real complex network.It is a network consists ofgroups of nodes within which links are dense but among which links are sparse.In this paper, the evolving network includenode, link and community growth and we apply the community size preferential attachment and strength preferentialattachment to a growing weighted network model and utilize weight assigning mechanism from BBV model.Theresulting network reflects the intrinsic community structure with generalized power-law distributions of nodes'degreesand strengths.
