This paper evaluates the performance of boosted decision trees for tagging b-jets. It is shown, using a Monte Carlo simulation of WH→lνqqˉ events that boosted decision trees outperform feed-forward neural networks. The results show that for a b-tagging efficiency of 60% the light jet rejection given by boosted decision trees is about 35% higher than that given by neural networks.
Cite
@article{arxiv.physics/0702041,
title = {Tagging heavy flavours with boosted decision trees},
author = {J. Bastos},
journal= {arXiv preprint arXiv:physics/0702041},
year = {2007}
}