Aggregated Recommendation through Random Forests

Algorithm 1

Random decision tree.

Input:    , condition attributes , decision attribute

Output:

Method:

(1) if     or     then

(2)  leaf = the distribution of d in the

(3)  return

(4) end if

(5) if  (all d of   are equal)  then

(6)  leaf = the distribution of d in the

(7)  return

(8) end if

(9)   is  a  copy  of  C

(10)   false

(11) while  (not    do

(12)  att = an attribute randomly selected from C

(13)  if    then

(14)     true

(15)   

(16)  end if

(17)  

(18)  if    then

(19)   

(20)  end if

(21) end while

(22) if  (not    then

(23)  leaf = the distribution of d in the

(24)  return

(25) end if

(26) Decision tree attribute for Root = att

(27) for  (each possible value   of att)  do

(28)  Add a new tree branch below Root, corresponding to .

(29)  Let   be the subset of   that have the value

(30)  if    then

(31)   leaf = the distribution of d in the

(32)   return

(33)  else

(34)   , ,

(35)  end if

(36) end for