IDX (Norton, 1989) is a TDIDT algorithm that selects the attribute that maximizes the following heuristic function:
We implemented IDX by modifying C4.5 so that it selects the attribute that maximizes (5).
C4.5 uses a greedy search strategy that chooses at each step the attribute with the highest information gain ratio. IDX uses a lookahead strategy that looks n tests ahead, where n is a parameter that may be set by the user. We did not implement this aspect of IDX. The lookahead strategy would perhaps make IDX more competitive with ICET, but it would also complicate comparison of the heuristic function (5) with the heuristics (3) and (4) used by EG2 and CS-ID3.