Polestar++ An Intelligent Routing Engine for National Wide Public Transportation

Public transportation plays a critical role in people's daily life. It has been proven that public transportation is more environmentally sustainable, efficient, and economical than any other forms of travel. However, due to the increasing expansion of transportation networks and more complex travel situations, people are having difficulties in efficiently finding the most preferred route from one place to another through public transportation systems for both intra-city and inter-city trips. To this end, in this paper, we present Polestar++ , a data-driven engine for intelligent and efficient public transportation routing. Specifically, we first propose a novel hierarchical public transportation graph (HPTG) to model both intra-city and inter-city public transportation in terms of various travel costs, such as time or distance. Then, we introduce a general route search algorithm coupled with an efficient station binding method for efficient route candidate generation. After that, we propose a two-pass route candidate ranking module to capture user preferences under dynamic travel situations. In particular, we propose two re-ranking models to decide the proper order of public routes: (1) a light-weight and explainable gradient boosting decision tree (GBDT) based model that integrates features from various urban data sources, and (2) a wide and deep learning (WDL) based model that automatically captures high order feature interactions from both inter-city and intra-city routes. Finally, experiments on two real-world data sets demonstrate the advantages of Polestar++ in terms of both efficiency and effectiveness. Indeed, in early 2019, Polestar++ has been deployed on Baidu Maps, one of the world's largest map services. To date, Polestar++ is servicing over 330 cities, answers over a hundred millions of queries each day, and achieves substantial improvement of user click ratio.