Research Of Constructing The Social Network Of Airline Customers

Recently, the analysis of the value of airline customer aims at the individual passenger. It treats every single passenger as an unrelated and isolated entity. In our real world, however,passengers live in the network group and they communicate and interact with each other.Based on the data of the PNR and PSR, this paper mainly studies the construction of the social network of airline customers, it has great significance to analyze our airline customers more comprehensively and accurately, its main contents are as follows:On the basis of analyzing the social network property of airline customers embedded in PNR and PSR, it proposes a method of constructing the social network of airline customers based on flight record of passenger. By preprocessing the PNR and PSR, the flight record of passenger is obtained. The RFM model is improved to analyze the individual value of passenger. The attributes the node contains, identification rules of the tie between nodes, the support count and confidence of one trade between passengers and the expression method of tie are proposed. The experiment on real data sets of airlines shows the feasibility and validity of the method.According to the parameters of network characterization, this paper analyzes structure characteristics of the connected passenger subnet in detail. The result indicates the connected passenger subnet has the characteristic of community structure and the small world. It also has the characteristic of high clustering coefficient, positive matching index between the two nodes’ degree and positive correlation between degree and closeness degree of node.Finally it analyses the application strategy of the social network of airline customers from measuring passenger’s value, identifying passenger roles, identifying communities,customer exploration, the evolution of the network group, detecting and attacking the criminal network.