Finally, as already mentioned in section 2.3.2, Cho et al. [18] combine Decision
Trees and Association Rule Mining in a web shop RS. In their system, associa-
tion rules are derived in order to link related items. The recommendation is then
computed by intersecting association rules with user preferences. They look for as-
sociation rules in different transaction sets such as purchases, basket placement, and
click-through. They also use a heuristic for weighting rules coming from each of the
transaction sets. Purchase association rules, for instance, are weighted higher than
click-through association rules.