IBBAS: A Visual Analytics System of Large-Scale Traffic Data for Bus Body Advertising

Abstract

Bus body advertising planners have been plagued for a long time for formulating solutions effectively and comparing different solutions efficiently, because there is no such decision support system. In this study, we attempt to use the state-of-the-art data mining and visualization techniques to solve this problem. Using the traffic data, we design a visual analytics system named IBBAS (Intelligent Bus Body Advertising System). This system can deal with two main challenges of the problem: new brand promotion and target group advertising. First, we propose two algorithms: DivideT and EstimateT. DivideT is used for the bus frequency division and EsitimateT is used for the bus arrival time estimation. Then, based on LDA topic model, we design an optimization algorithm to generate bus body advertising solution based on multiple constraints. Finally, we use the visualization techniques to create a system. There are some functional modules used in the system: the heatmap changing with time displays the passenger flow volume of each bus station; the candidate panel allows users to expediently modify the original solution and forms a tailored solution; the compare panel clearly uncovers the differences among multiple solutions with a flexible ranking tool. This system has been demonstrated using case studies with a real-world dataset. We also collected feedback from domain experts and conducted a preliminary evaluation criteria.