Dark Photon Search at BABAR

Greenwood, Ross N ; /MIT /SLAC

关键词: EFFICIENCY;    LEARNING;    MASS RESOLUTION;    MULTIVARIATE ANALYSIS;    PHOTONS;    RESONANCE;    SENSITIVITY;    TRAINING Experiment-HEP;    OTHER;   

DOI  :  10.2172/1050217
RP-ID  :  SLAC-TN-12-023
PID  :  OSTI ID: 1050217
Others  :  TRN: US1204686

学科分类：物理（综合）

美国|英语

来源: SciTech Connect

【 摘 要 】

Presented is the current progress of a search for the signature of a dark photon or new particle using the BaBar data set. We search for the processes e{sup +}e{sup -} {yields} {gamma}{sub ISR}A{prime},A{prime} {yields} e{sup +}e{sup -} and e{sup +}e{sup -} {yields} {gamma}{sub ISR}{gamma}, {gamma} {yields} A{prime},A{prime} {yields} e{sup +}e{sup -}, where {gamma}{sub ISR} is an initial state radiated photon of energy E{sub {gamma}} >= 1 GeV. Twenty-five sets of Monte Carlo, simulating e{sup +}e{sup -} collisions at an energy of 10.58 GeV, were produced with different values of the A{prime} mass ranging from 100 MeV to 9.5 GeV. The mass resolution is calculated based on Monte Carlo simulations. We implement ROOT's Toolkit for Multivariate Analysis (TMVA), a machine learning tool that allows us to evaluate the signal character of events based on many of discriminating variables. TMVA training is conducted with samples of Monte Carlo as signal and a small portion of Run 6 as background. The multivariate analysis produces additional cuts to separate signal and background. The signal efficiency and sensitivity are calculated. The analysis will move forward to fit the background and scan the residuals for the narrow resonance peak of a new particle.