Optimization of Signal Significance by Bagging Decision Trees

An algorithm for optimization of signal significance or any other classification figure of merit (FOM) suited for analysis of HEP data is described. This algorithm trains decision trees on many bootstrap replicas of training data with each tree required to optimize the signal significance or any other chosen FOM. New data are then classified by a simple majority vote of the built trees. The performance of the algorithm has been studied using a search for the radiative leptonic decay B → γlν at BABAR and shown to be superior to that of all other attempted classifiers including such powerful methods as boosted decision trees. In the B → γeν channel, the described algorithm increases the expected signal significance from 2.4σ obtained by an original method designed for the B → γlν analysis to 3.0σ.