In many applications that involve identification of an unknown monotone Boolean function (MBF), the cost of inferring the value of a vector using monotonicity is negligible compared to the cost of querying its value. Accordingly, heuristics to identify MBFs seek to minimize the number of vectors queried. The order in which queried vectors are selected determines the number of queries needed. This paper presents a method for MBF identification that iteratively selects vectors to be queried based on a regret-minimization criterion. We observe that the best extant heuristic may be considered a special case of our approach and present alternate regret functions that perform no worse than this heuristic.