Algorithm 1 Pseudo Code for ODAELM-S.

Input:

$L:=$ the number of hidden layer neurons;

$Act:=$ the activation function type;

$SD:=$ the source domain data;

1: Initialize two empty sets, i.e., $LSet$ and $USet$, as labeled and unlabeled sets, respectively;

2: Set activation function as $Act$ and initialize an ELM with L hidden nodes with $SD$;

3: Let $Hs$ be defined as in Equation (9);

4: while new sample x in the target domain arrives do

5: Calculate the probability P for labeling;

6: Generate random value between 0 and 1 as p;

7: if $p<P$ then

8:  Add x to $USet$;

9:  Select a group of samples from $USet$ as $\delta Set$ for labeling;

10:  if $H_S$ has more rows then

11:   Update the classifier using Equations (11) and (12);

12:  else

13:   Update the classifier using Equation (14) and (15);

14:  end if

15:  Set $LSet=LSet-\delta Set$ and $USet=USet-\delta Set$;

16: else

17:  Add x to $USet$;

18: end if

19: end while