Background

The industry of transportation is undergoing profound changes with the application of artificial intelligence（AI）, big data analysis, cloud computing and other new technologies. The UAI Data Contest is designed to improve efficiency and accuracy of calculation on big data application platforms using real-life traffic data provided, thus to accelerate the development of an intelligent transportation system.

UCAR Group, together with IEEE China and IEEE Computer Society, launches the first UAI Data Contest now open for application by AI experts and enthusiasts from around the world. The contest has been designed by the UCAR technical team. Contestants will be given historical UCAR order information based on which they are expected to forecast the total number of orders for rides leaving from one specified zone heading to another in a certain period in the future.
 

Company Background

UCAR is a large company group with its business deeply focuses on developing intelligent transportation and a whole industry chain for automobiles. It’s three major business divisions are transportation (CAR Inc. & Shenzhou Zhuanche), automobile e-commerce (Shenzhou Maimai Che) and automobile finance (Shenzhou Shandai). The other areas it has also entered include Car on the Internet, consumer finance, and Internet insurance.

Of the main divisions, CAR Inc. is a car rental service founded in 2007. Targeting mainly at those prefer to rent and drive cars for short-distance travels, it has grown into the largest car rental chain in China. Shenzhou Zhuanche is a high-end online ride-hailing service which is committed to provide safe and cozy rides and standardized services with its own fleet and professionally trained chauffeurs. The one-stop automobile e-commerce platform Shenzhou Maimaiche has covered 123 cities already. And Shenzhou Shandai, the financing service, in 114 cities.

Utilizing massive real-world data generated from the above-said three divisions and cutting-edge technologies such as big data, cloud computing and AI, UCAR has been able to develop an original man-car ecology intelligent system called UCAR Brain. It offers not only technical support for an efficient and coordinated growth of the different divisions but also data models and solutions for the construction of an intelligent transportation system. In August UCAR launched the UCAR smart Transportation Open Platform (UTOP), making public its 10-year real-world data resources and an AI technology platform to facilitate developers and to push forward innovation in the fields of intelligent transportation and smart cities.

Task Description

At any moment, a user can open the Shenzhou Zhuanche APP, enter a pickup point and a destination, and order a ride. Whether it gets answered or not, we regard such an action as a request for a ride from location A to location B.

Contestants will be expected to predict, based on order data given by Shenzhou Zhuanche, the volume of orders for rides from zone A to zone B during a certain period of time. (The orders include both successful and failed ones, i.e., the total number of “requests for ride”). The training and test data are collected from selected orders in a single city in July and August 2017.