Identification of b-quark jets with the CMS experiment (1211.4462v2)

Abstract: At the Large Hadron Collider, the identification of jets originating from b quarks is important for searches for new physics and for measurements of standard model processes. A variety of algorithms has been developed by CMS to select b-quark jets based on variables such as the impact parameters of charged-particle tracks, the properties of reconstructed decay vertices, and the presence or absence of a lepton, or combinations thereof. The performance of these algorithms has been measured using data from proton-proton collisions at the LHC and compared with expectations based on simulation. The data used in this study were recorded in 2011 at sqrt(s) = 7 TeV for a total integrated luminosity of 5.0 inverse femtobarns. The efficiency for tagging b-quark jets has been measured in events from multijet and t-quark pair production. CMS has achieved a b-jet tagging efficiency of 85% for a light-parton misidentification probability of 10% in multijet events. For analyses requiring higher purity, a misidentification probability of only 1.5% has been achieved, for a 70% b-jet tagging efficiency.

• The paper presents novel algorithms combining impact parameter and secondary vertex data to effectively tag b-quark jets.
• It demonstrates that the CSV algorithm achieves 85% efficiency with a 10% misidentification probability in multijet events.
• Systematic uncertainties and derived scale factors ensure robust corrections between data and simulation for precise analyses.

Identification of b-Quark Jets with the CMS Experiment

The paper "Identification of b-quark jets with the CMS experiment" delineates the procedures and outcomes of employing various algorithms for b-jet tagging at the Large Hadron Collider (LHC), specifically within the CMS experiment. The ability to accurately identify jets originating from b quarks (\cPqb\ jets) is of substantial importance for probing new physics and for precise measurements of standard model phenomena such as the decays involving top quarks or the Higgs boson.

Overview of Algorithms

Numerous algorithms are developed by the CMS collaboration to efficiently tag b-jets. These algorithms utilize a variety of discriminative features:

• Impact Parameters of Tracks: The signed impact parameter significance is pivotal for distinguishing b-jet decay products from other particles.
• Secondary Vertex Characteristics: The properties of secondary vertices, such as flight distance significance and vertex mass, are utilized in tagging.
• Combined Approaches: Advanced techniques like the Combined Secondary Vertex (CSV) algorithm leverage both track-based and vertex-based information for enhanced discrimination.

Each b-tagging algorithm provides selections at different operating points — loose, medium, and tight — representing trade-offs between b-jet efficiency and misidentification rates (for light-parton jets).

Measured Performance

Quantitatively, the efficiency and misidentification probability of these algorithms were evaluated through different event samples. For instance, the CSV algorithm achieves an 85% b-jet tagging efficiency at a 10% misidentification probability for light-parton jets in multijet events.

1. Multijet Events: Various methods like PtRel, System8, and LT are applied to multijet datasets enriched with muons to determine the b-jet tagging efficiency across a wide range of transverse momenta (30-670 GeV).
2. \ttbar Events: The efficiency measurements in \ttbar events use Profile Likelihood Ratio (PLR) and Flavour Tagging methods which align with expected \cPqb-jet efficiencies in these events.
3. Misidentification Probability: Using negative taggers, the probability of light-parton jets being misidentified as b-jets is quantified, providing essential corrections to be applied to simulations.

Systematic Uncertainties and Scale Factors

The paper meticulously accounts for systematic uncertainties from sources such as pileup, gluon splitting, and jet energy scale variations, ensuring robust measurements. Derived scale factors are provided to correct for discrepancies between data and simulation, enabling their application in diverse physics analyses conducted by the CMS collaboration.

Implications

The outcomes of this work have far-reaching implications for high-energy physics. They enable precision studies of processes involving heavy-flavour jets, and they are integral in the exploration of phenomena beyond the Standard Model. Future advancements in AI and machine learning approaches could further refine b-jet tagging algorithms, potentially enhancing pattern recognition capabilities and real-time data processing at the LHC.

Overall, the document serves as a foundational reference for CMS analyses requiring accurate b-jet identification, substantiating the robustness of the CMS experiment's jet-tagging strategy in exploring the intricacies of particle physics.