Comparison of the use of binary decision trees and neural networks in top-quark detection

Bowser-Chao, David; Dzialo, Debra Lynn

doi:10.1103/physrevd.47.1900

Cited by 22 publications

(12 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We find an optimum performance with the gradient boosted decision tree classifier (BDT) which offers the advantage over neural networks that the use of variables that do not discriminate significantly between a particular signal and background, or are highly correlated with other variables, does not compromise the classifier performance. In the BDT [34,35] approach, a series of splittings of the event sample is made at a sequential set of nodes into background-like and signal-like subsample nodes. The splitting is based upon the purity of signal and background events in a given node N and its signal-like and background-like daughter nodes S and B.…”

Section: Multivariate Analysismentioning

confidence: 99%

Search for the Higgs boson in lepton, tau, and jets final states

Abazov¹,

Abbott²,

Acharya³

et al. 2013

Phys. Rev. D

View full text Add to dashboard Cite

We present a search for the standard model Higgs boson in final states with an electron or muon and a hadronically decaying tau lepton in association with two or more jets using 9:7 fb À1 of Run II Fermilab Tevatron Collider data collected with the D0 detector. The analysis is sensitive to Higgs boson production via gluon fusion, associated vector boson production, and vector boson fusion, followed by the Higgs boson decay to tau lepton pairs or to W boson pairs. The ratios of 95% C.L. upper limits on the cross section times branching ratio to those predicted by the standard model are obtained for orthogonal subsamples that are enriched in either H ! decays or H ! WW decays, and for the combination of these subsample limits. The observed and expected limit ratios for the combined subsamples at a Higgs boson mass of 125 GeV are 11.3 and 9.0, respectively.

show abstract

Section: Multivariate Analysismentioning

confidence: 99%

Search for the Higgs boson in lepton, tau, and jets final states

Abazov¹,

Abbott²,

Acharya³

et al. 2013

Phys. Rev. D

View full text Add to dashboard Cite

show abstract

“…The boosted decision tree (BDT) multivariate analysis technique [39][40][41] is also used to distinguish between the W signal and the background. For the BDT analysis we apply all the selection criteria described in Sec.…”

Section: The Boosted Decision Tree Analysismentioning

confidence: 99%

Search for a W′ boson decaying to a bottom quark and a top quark in pp collisions at s=7 T

Chatrchyan¹,

Khachatryan²,

Sirunyan³

et al. 2013

Physics Letters B

View full text Add to dashboard Cite

Results are presented from a search for a W boson using a dataset corresponding to 5.0 fb −1 of integrated luminosity collected during 2011 by the CMS experiment at the LHC in pp collisions at √ s = 7 TeV. The W boson is modeled as a heavy W boson, but different scenarios for the couplings to fermions are considered, involving both lefthanded and right-handed chiral projections of the fermions, as well as an arbitrary mixture of the two. The search is performed in the decay channel W → tb, leading to a final state signature with a single electron or muon, missing transverse energy, and jets, at least one of which is identified as a b-jet. A W boson that couples to the right-handed (left-handed) chiral projections of the fermions with the same coupling constants as the W is excluded for masses below 1.85 (1.51) TeV at the 95% confidence level. For the first time using LHC data, constraints on the W gauge couplings for a set of left-and right-handed coupling combinations have been placed. These results represent a significant improvement over previously published limits.

show abstract

“…A decision tree [57,58] employs a machine-learning technique that effectively extends a simple cut-based analysis into a multivariate algorithm with a continuous discriminant output. Boosting is a process that can be used on any weak classifier (defined as any classifier that does a little better than random guessing).…”

Section: Boosted Decision Trees Analysismentioning

confidence: 99%

Evidence for production of single top quarks

Abazov¹,

Abbott²,

Abolins³

et al. 2008

Phys. Rev. D

View full text Add to dashboard Cite

We present first evidence for the production of single top quarks in the D0 detector at the Fermilab Tevatron ppbar collider. The standard model predicts that the electroweak interaction can produce a top quark together with an antibottom quark or light quark, without the antiparticle top quark partner that is always produced from strong coupling processes. Top quarks were first observed in pair production in 1995, and since then, single top quark production has been searched for in ever larger datasets. In this analysis, we select events from a 0.9 fb-1 dataset that have an electron or muon and missing transverse energy from the decay of a W boson from the top quark decay, and two, three, or four jets, with one or two of the jets identified as originating from a b hadron decay. The selected events are mostly backgrounds such as W+jets and ttbar events, which we separate from the expected signals using three multivariate analysis techniques: boosted decision trees, Bayesian neural networks, and matrix element calculations. A binned likelihood fit of the signal cross section plus background to the data from the combination of the results from the three analysis methods gives a cross section for single top quark production of 4.7 +- 1.3 pb. The probability to measure a cross section at this value or higher in the absence of signal is 0.014%, corresponding to a 3.6 standard deviation significance. The measured cross section value is compatible at the 10% level with the standard model prediction for electroweak top quark production.Comment: 51 pages, 30 figures. Updated version after review and acceptance in PR

show abstract

Comparison of the use of binary decision trees and neural networks in top-quark detection

Cited by 22 publications

References 7 publications

Search for the Higgs boson in lepton, tau, and jets final states

Search for the Higgs boson in lepton, tau, and jets final states

Search for a W′ boson decaying to a bottom quark and a top quark in pp collisions at s=7 T

Evidence for production of single top quarks

Contact Info

Product

Resources

About