Multisource data integration for targeted bus exterior advertising

Bus exterior advertising plays a significant role in outdoor advertising, since it provides frequent exposure to large number of residents. Traditional route selection methods are generally based on rough estimation, for example, the total passengers bus or geographical features along route. Targeted remains challenge as little is known about characteristics people In this study, we aiming at determining set routes given ad category maximize effectiveness, by mining multiple data sources, including mobile phone data, GPS smart card (SCD), and land use data. Specifically, first estimated distribution potential target audiences using Two optimization models proposed considering different requirements. For well-established brands that familiar with, wide coverage-oriented model coverage audiences. new require high level before they become recognizable, deep times ads. Both were demonstrated with case study Shenzhen, China explicitly present outcomes differences between them. The calculation results show achieves an average 84.8% improvement compared baseline 1 which selects most passengers, while 9.2% 2 maximum area reaching more intensity almost 3.7 model. provide options advertisers select suitable strategy according their needs.