Abstract Constraint acquisition systems assist the non-expert user in modeling her problem as a constraint network. Most existing constraint acquisition systems interact with the user by asking her to classify an example as positive or negative. Such queries do not use the structure of the problem and can thus lead the user to answer a large number of queries. In this paper, we propose P REDICT &A SK, an algorithm based on the prediction of missing constraints in the partial network learned so far. Such missing constraints are directly asked to the user through recommendation queries, a new, more informative kind of queries. P REDICT &A SK can be plugged in any constraint acquisition system. We experimentally compare the Q UACQ system to an extended version boosted by the use of our recommendation queries. The results show that the extended version improves the basic Q UACQ.