{
  "fetched_at": "2026-10-24T16:12:06Z",
  "source_url": "https://www.kaggle.com/competitions/flavour-tagging/data",
  "pages": [
    {
      "name": "Description",
      "content": "<p>Goal of this competition is to determine the flavour of charged B mesons (mesons that contain <strong>b</strong> or<strong> anti-b</strong> quark).</p>\r\n<p>In simpler words, guess whether particle is B+ or B- (guess, whether the particle is meson or anti-meson). </p>\r\n<p>This is the second challenge in the course of machine learning organized by <a href=\"http://yandexdataschool.com\">Yandex School of Data Analysis</a> at <a href=\"https://www.imperial.ac.uk\">Imperial College London</a>.</p>\r\n<h2>Idea of tagging</h2>\r\n<p>Mesons are particles that contain a quark and an anti-quark. <a href=\"https://en.wikipedia.org/wiki/B_meson\"><strong>B</strong> mesons</a> are mesons that contain <strong>b</strong> or <strong>anti-b</strong> quark.</p>\r\n<p>Each B-meson production is accompanied by birth of other particles that leave tracks in the detector. Type and properties of particles and their corresponding tracks are tightly connected with a quark (b or anti-b) forming a meson of interest.</p>\r\n<p>Thus, we can guess the quark in the B-meson by analyzing properties of tracks in the detector.</p>\r\n<p>Bad news: there are many tracks in the detector, some may have no connection to B-meson. Thus, it is a problem with noise.</p>\r\n<p>However, the problem is interesting, because we <a href=\"https://en.wikipedia.org/wiki/Color_confinement\">cannot</a> observe a single quark, but only a hadron (particle consisting of several quarks). Thus, such techniques potentially give us ability to study the quark properties.</p>\r\n<p>No, we don't assume that you know physics. </p>\r\n<h2>Data Files</h2>\r\n<p><strong>trainingB.root</strong> - training samples with B-meson description<br /><strong>trainingTrack.root</strong> - training samples with other particles in a collision<br /><strong>testB.root</strong> -  test samples with B-meson description<br /><strong>testTrack.root</strong> -  test samples with other particles in a collision<br /><strong>sample_submission.csv</strong> - simple submission, demonstrates expected format</p>\r\n<h2>Data Format</h2>\r\n<p><strong>trainingB.root</strong> and <strong>testB.root</strong> contain information about the B meson (goal: predict B-meson charge). Column <strong>chargeB</strong> is a target.</p>\r\n<p><strong>trainingTracks.root</strong> and <strong>testTracks.root</strong> contain information about the tracks in the collision. Tracks with the same <strong>ID</strong> corresponds to the same collision and the B-meson corresponding to this collision is in the <strong>trainingB.root</strong> and <strong>testB.root</strong> with the same <strong>ID</strong>.</p>\r\n<p>The file for B-meson looks like:</p>\r\n<pre>ID ptB etaB phiB chargeB<br />0 7.487484 3.598782 2.042920 1<br />1 1 6.157421 2.812284 4.060636 1</pre>\r\n<p>The file for tracks looks like (both tracks connected to the same B-meson with ID=0):</p>\r\n<pre>ID ptTrack etaTrack phiTrack track_quality PIDNNe ...<br />0 0.768184 2.004773 0.183600 1.357150 7.140722e-08 ...<br />0 0.389962 2.694220 0.216586 2.021095 1.792256e-06 ...</pre>\r\n<p>The test samples have the same structure (with the exception of <strong>chargeB</strong> column which is absent in test). </p>\r\n<h3>Features Description</h3>\r\n<p>B-mesons features:</p>\r\n<p><strong>ID</strong> - collision ID</p>\r\n<p><strong>ptB </strong>- transverse momentum of B-meson</p>\r\n<p><strong>etaB</strong> - <a href=\"https://en.wikipedia.org/wiki/Pseudorapidity\">pseudorapidity</a> of B-meson. It is almost the same as polar angle, see wikipedia</p>\r\n<p><strong>phiB</strong> - azimuth angle of B-meson</p>\r\n<p><strong>chargeB</strong> - the target (is absent in test samples). -1 corresponds to b-quark and 1 correspond to anti-b quark.</p>\r\n<p>For every track registered in the event we calculate several features. Each track has following description:</p>\r\n<p><strong>ID</strong> - collision ID</p>\r\n<p><strong>ptTrack</strong> - transverse momentum</p>\r\n<p><strong>etaTrack</strong> - pseudorapidity </p>\r\n<p><strong>phiTrack</strong> - azimuth angle</p>\r\n<p><strong>track_quality</strong> - quality of track reconstruction</p>\r\n<p><strong>PIDe</strong> - particle identification: probability for this track to come from electron</p>\r\n<p><strong>PIDk</strong> - particle identification: probability for this track to come from kaon</p>\r\n<p><strong>PIDmu</strong> - particle identification: probability for this track to come from muon</p>\r\n<p><strong>PIDpi</strong> - particle identification: probability for this track to come from pion</p>\r\n<p><strong>PIDp</strong> - particle identification: probability for this track to come from proton</p>\r\n<p><strong>PIDghost</strong> - particle identification: probability to be a ghost track (= noise, which was erroneously considered as a track left by particle)</p>\r\n<p><strong>IP</strong> - impact parameter: the distance between the track and vertex from which B-meson appeared</p>\r\n<p><strong>IPerr</strong> - the error of the determination of IP</p>\r\n<p><strong>vertexIP</strong> - impact parameter: the minimum distance between the track and other reconstructed collision vertices.</p>\r\n<p><strong>EOverP</strong> - E/P, where E - energy deposit in the calorimeter system, P - momentum of the particle</p>\r\n<p><b>Edx</b> - energy deposit per dx at the beginning of track</p>\r\n<p><strong>chargeTrack</strong> - charge of a track</p>\r\n<h2>Discussions</h2>\r\n<p>Please discuss the competition in the <a href=\"https://gitter.im/MLatImperial2017/Lobby\">gitter</a> chat of the course.  </p>\r\n<h2>Submission Format</h2>\r\n<pre>ID,Prediction<br />0,0.54<br />1,0.34<br />2,0.22<br />...</pre>\r\n<p>The higher predictions should correspond to +1 B-meson charge and lower to -1.</p>\r\n<h2>Evaluation</h2>\r\n<p>Metric computed on the leaderboard is area under the <a href=\"http://arogozhnikov.github.io/RocCurve.html\">ROC curve</a> (ROC AUC). This metric is discussed in the course.</p>\r\n<p>Validation dataset is split in 2 equal parts: public and private. First is shown on the leaderboard, second is used to compute final standings.</p>\r\n<h2>Getting started</h2>\r\n<p>The code to start will be published in the <a href=\"https://github.com/yandexdataschool/MLatImperial2017\">repository of the course</a>.</p>\r\n<h2>Teams</h2>\r\n<p>You can participate in the teams of two!</p>\r\n<p>Only <strong>one</strong> student from physics department can be in the team.</p>\r\n<h2>Acknowledgements</h2>\r\n<p>We are grateful to the LHCb collaboration at LHC (and personally to Guy Wilkinson) for allowing us to use simulation samples in this competition. Thanks to Denis Derkach for preparing data samples.</p>\r\n<h2>Disclaimer</h2>\r\n<p>Provided samples were preprocessed for the purposes of the competition. No judgment can be done based on these samples about the properties of LHCb detector.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "\r\n\t<!-- Begin Base Rules -->\r\n    <h3>One account per participant</h3>\r\n    <p>You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.</p>\r\n    <h3>No private sharing outside teams</h3>\r\n    <p>\r\n        Privately sharing code or data outside of teams is not permitted.\r\n        It's okay to share code if made available to all participants on the forums.\r\n    </p>\r\n    <h3>Team Mergers</h3>\r\n    <p>Team mergers are not allowed in this competition.</p>\r\n\r\n    <h3>Team Limits</h3>\r\n    <p>The maximum size of a team is 2 participants.</p>\r\n    <h3>Submission Limits</h3>\r\n    <p>You may submit a maximum of 15 entries per day.</p>\r\n    <p>You may select up to 2 final submissions for judging.</p>\r\n    <h3>Competition Timeline</h3>\r\n    <p>Start Date: <strong>1/25/2017 10:16 PM UTC</strong></p>\r\n    <p>Merger Deadline: <strong>None</strong></p>\r\n    <p>Entry Deadline: <strong>None</strong></p>\r\n    <p>End Date: <strong>2/5/2017 7:59 PM UTC</strong></p>\r\n\t<!-- End Base Rules -->\r\n<p>INTELLECTUAL PROPERTY</p>\r\n<p>DATA</p>\r\n<p>'Data' means the Data or Datasets linked from the Competition Website for the purpose of use by Participants in the Competition. For the avoidance of doubt, Data is deemed for the purpose of these Competition Rules to include any prototype or executable code provided to Participants by Kaggle or Competition Organizer via the Website. Participants must use the Data only as permitted by these Competition Rules and any associated data use rules specified on the Competition Website.</p>\r\n<p>Unless otherwise permitted by the terms of the Competition Website, Participants must use the Data solely for the purpose and duration of the Competition, including but not limited to reading and learning from the Data, analyzing the Data, modifying the Data and generally preparing your Submission and any underlying models and participating in forum discussions on the Website. Participants agree to use suitable measures to prevent persons who have not formally agreed to these Competition Rules from gaining access to the Data and agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Data to any party not participating in the Competition. Participants agree to notify Kaggle immediately upon learning of any possible unauthorized transmission or unauthorized access of the Data and agree to work with Kaggle to rectify any unauthorized transmission. Participants agree that participation in the Competition shall not be construed as having or being granted a license (expressly, by implication, estoppel, or otherwise) under, or any right of ownership in, any of the Data.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}