Discovering And Analyzing Family Groups In Passenger Social Networks

With the development of information technology, all walks of life can store and utilize massive data effectively, so business system of passenger transport industry has accumulated the data of passenger information and travel behavior. At the same time, people travel more and more frequently and choose a variety of ways to travel with the development of economy and society. People usually travel together with others in groups, such as with family members for visiting relatives, with colleagues for business, with friends for sightseeing and so on. Especially, the family groups, as a kind of the most common consumer units, have a considerable scale in the field of passenger transportation market. If we can effectively utilize data of business system to accurately identify family groups, it will help the carriers provide passengers with personalized travel services and product recommendations, improved service quality, reduced service cost and improved satisfaction.This paper studies the issue of finding family groups in the field of civil aviation and proposes a family group detection method based on passenger social networks. First of all, we construct passenger social networks based on their co-travel behaviors extracted from the historical travel records; then, on the basis of co-travel networks, we generate three types of features (historical co-travel statistics, demographic characteristics and network-based features) to express the different relationship between passengers. We use a collective classification algorithm to classify the social relationships between passengers (family or non-family relationship); finally, we employee a weighted community detection algorithm to find family groups, which takes the relationship classification results as weights. Finally, we analyze the travel behavior characteristics on the discovered family groups.Experimental results on a real data set of passenger travel records in the field of civil aviation demonstrate that our method can effectively find family groups from historical travel records, and then analyze the travel behavior of family groups. Discovering family groups will help aviation enterprises with precision marketing or personalized service; and analyzing family groups can provide decision support for passenger transport industry.