Data description:
This package contains two traffic data streams were generated based on traffic flow from two China National Highways (URL£ºhttp://traflow.fudan.edu.cn/data.html), 
namely G2 (Beijing-Shanghai) and G60 (Shanghai-Kunming).
(1) For TrafficStream 1, it consists of 44,033 instances and three numerical attributes: velocity, acceleration, and location, 
which were collected from G2 and G60. Meanwhile, there are three classes to evaluate, and the class labels 0, 1 and 2 represent 
cars, trucks and unknown, respectively.
(2) For TrafficStream 2, it is also a collection of 24,103 vehicles from G2 and G60. Each sample in TrafficStream 2 denotes a 
car (marked as 0) or truck (tagged as 1) with three different features: velocity, acceleration, and location.

TrafficStream 1£º
#Instance : 44,033
#Feature: 3
 #Class: 3

TrafficStream 2£º
#Instance : 24,103
#Feature: 3
 #Class: 2

NOTE: Scholars can feel free to use these two datasets (for academic purpose only) at your own risk. 
For other purposes, please contact Dr. Yunlong Mi (miyunlong17@mails.ucas.ac.cn). 

NOTE2: Please cite our work if you use these data in any way, as follows:
Y. L. Mi. ConceptNN: A concept neural network for dynamic stream learning, 2023, 
submitted to IEEE TPAMI for consideration.