Algorithm 1. RAQ
Input:	A dataset S with features: Fmt, Fmh, Fmp, Fmw, Fmv, Fri, Ftcs, Fpn and labeled AQI level; unlabeled dataset U; trees quantity T; features quantity m;
Output:	AQI level
1	for T trees
2	randomly select m features from S;
3	for m features in each node
4	calculate information gain by Equation (3);
5	choose maximum gain to split the dataset in the node;
6	remove used feature from feature candidates;
7	input unlabeled data into trees;
5	get predicted AQI level according to Equations (5) and (6);