About adaboost algorithm

Question

I’m working on traffic forecasting, where I can predict that a place has heavy or light traffic. I classified each traffic as 1-5, 1 the lightest traffic, and 5 the heaviest traffic.

I came across this site http://www.waset.org/journals/waset/v25/v25-36.pdf , the AdaBoost algorithm, and I really have difficulty learning this algorithm. Especially in the part where S is the set (( xi , yi ), i=(1,2,…,m) ). where Y={-1,+1} . What is x , y and constant L ? What is the meaning of L ?

Can someone explain this algorithm to me? :)

Answer 1

S={(x1,y1),...,(xm,ym)} : Each pair (x,y) is a pattern used to train (or test) your classifier:

• x = Functions that describe this particular pattern, for example, values ​​that list amount of cars on the road , day of the week , etc.
• y = Label for a specific x , which in your case may be 1, 2, 3, 4 or 5

Table 1 in the document shows the x functions that they used, namely: DAY , TIME , INT , DET , LINK , POS , GRE , DIS >, VOL and OCC . The last column of the table shows the label ( y ), which they set to either 1 or -1 (i.e. yes or no ). Each row in the table is 1 sample.

L - the number of rounds in which AdaBoost teaches a weak student (in the article Random Forests used as a weak classifier). If you set L to 1 , then AdaBoost will run 1 round, and only one weak classifier will train, which will have poor results. Perform several experiments with different values ​​for L to find the optimal value (i.e. when AdaBoost converges or when it starts to rewrite).