Algorithm 2 Algorithm for induction of decision rules.

Input: decision table T with numerical values of attributes, number p of best attributes to be taken into consideration

Output: set of unique rules R

$j\leftarrow \varnothing$;

$Q\leftarrow \varnothing$;

convert T into $EAV$ table;

${\forall }_{attr\in A}$ calculate $ST{D}_{attr}$ grouped by $V_d$ and create a ranking;

select p attributes from the ranking and select descriptors from $AV$ containing selected attributes;

while all selected descriptors are not processed do

create separable subtable $T^j(attr,v_{attr})$;

$Q\leftarrow Q\cup \{attr=v_{attr}\}$;

if $T^j(attr,v_{attr})$ is degenerate OR $j=p$ then

$R\leftarrow R\cup {\forall }_{attr=v_{attr}\in Q}(att{r}_i=v_{att{r}_i})\to mcd\left(T^j\right)$, where $mcd\left(T^j\right)$ is the most common decision for $T^j$;

else

$j=j+1$;

end if

end while