Data
COMET_MC

COMET_MC

active ARFF Attribution-ShareAlike (CC BY-SA) Visibility: public Uploaded 17-03-2016 by Nikita Kazeev
0 likes downloaded by 0 people , 0 total downloads 0 issues 0 downvotes
  • Economics Social Media study_10
Issue #Downvotes for this reason By


Loading wiki
Help us complete this description Edit
Author: COMET collaboration Acknowledgements: Chen WU, Ewen Gillies Source: Unknown - Date unknown Please cite: Monte-Carlo simulation of COMET detector, COMET collaboration, http://comet.kek.jp/ ## Guess which points belong to signal track [COMET](http://comet.kek.jp/Introduction.html) is an experiment being constructed at the J-PARC proton beam laboratory in Japan. It will search for coherent neutrino-less conversion of a muon to an electron, μ- + N(A,Z) → e- + N(A,Z). This process breaks the law of lepton conservation. If detected, it will be a signal of new physics. The previous upper limit for this decay was set [5] by the SINDRUM II experiment in 2006. COMET is designed to have 10,000 times better sensitivity. Wires positions are available in a [supplementary file](https://drive.google.com/file/d/0B_gdsqrqzUJyMHcyUVFHa05FLXc/view?usp=sharing) ## Cylindrical Drift Chamber The COMET experiment is looking for muon to electron conversion, μ- + N → e- + N. COMET Phase-I will the Cylindrical Drift Chamber as the primary detector for physics measurements. Specifically, the momentum of resulting particles will be measured using the CyDet, which is a cylindrical wire array detector. The particles flying out of muon-stopping target and registered by the CyDet. Among those we are interested in tracks left by electrons with specific energy, which are produced by muon to electron conversion. The CyDet consists of 4482 sensitive wires organized in 18 layers. Each wire measures the energy deposited by a passing charged particle. Within each of the layers, the wires have same distance to the stopping target and stereometry angle. ![Scheme of COMET cylindrical detector](https://kaggle2.blob.core.windows.net/competitions/inclass/4520/media/comet_3d.gif) There is magnetic field in the detector, which causes electron moves in helical path as shown below. This electron deposits energy in the wires close to the flight path. The radius of helix is proportional to transverse momentum of the electron: R = p_t/(eB) where p_t is transverse momentum, B is strength of magnetic field, e is charge of electron. ![Trajectory of electron in margetic field](https://kaggle2.blob.core.windows.net/competitions/inclass/4520/media/COMEThelixing.png) The energy deposited on each wire is measured at the end plate of the cylindrical detector. An example of the resulting signal event can be seen below, where blue dots are background hits and red are hits from signal electrons: ![Energy depositions in COMET](https://kaggle2.blob.core.windows.net/competitions/inclass/4520/media/COMET2dprojection.png) ## Data format Format of data Information about position of straw tubes is kept in [supplementary file](https://drive.google.com/file/d/0B_gdsqrqzUJyMHcyUVFHa05FLXc/view?usp=sharing), it has following information: 1. wire_id - zero-based index of straw tubes 2. wire_rho - distance to stopping target 3. wire_phi - phi angle in the plane perpendicular to beam Each event is a 'snapshot' of detector, it consists of data taken from all 4482 wires: energy and relative time. Event is written only when hodoscope detects particle. Relative time is difference between time when straw tube detected energy and time of hodoscope. Data in train.csv is organized as following: 1. event_id - zero-based index of event 2. wire_id - zero-based index of wire 3. global_id - zero-based identifier for each entry in file 4. energy - energy deposited at wire 5. relative_time = ( time of readout - hodoscope time) 6. label (0 = not activated or background, 1=signal hit) Wire with zero energy deposition is not activated. ## More details 1. [COMET official site](http://comet.kek.jp/) 2. [COMET conceptual design report](http://comet.kek.jp/Documents_files/comet-cdr-v1.0.pdf) 3. [Раритеты микромира](https://nplus1.ru/news/2015/05/29/reareevents) - if you aren't deep into HEP, this article in Russian is probably good starting point to understand what is COMET about. 4. [COMET presentation](http://www-physics.lbl.gov/seminars/old/LBNL2014KUNO.pdf) 5. [A search for μ-e conversion in muonic gold](http://www.researchgate.net/publication/226763791_A_search_for_-e_conversion_in_muonic_gold) ## Important note Datasets available for this challenge are results of preliminary Monte Carlo simulation. They don't completely represent properties of COMET's detector and thus cannot be used to estimate final properties of tracking system, but are appropriate to test different approaches to tracking. ## Acknowledgements We thank COMET collaboration (and specially Chen WU) for allowing us to use this dataset.

6 features

label (target)numeric2 unique values
0 missing
global_id (row identifier)numeric7619400 unique values
0 missing
event_idnumeric1700 unique values
0 missing
wire_idnumeric4482 unique values
0 missing
energy_depositnumeric167061 unique values
0 missing
relative_timenumeric1171897 unique values
0 missing

107 properties

7619400
Number of instances (rows) of the dataset.
6
Number of attributes (columns) of the dataset.
0
Number of distinct values of the target attribute (if it is nominal).
0
Number of missing values in the dataset.
0
Number of instances with at least one value missing.
6
Number of numeric attributes.
0
Number of nominal attributes.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 3
Kappa coefficient achieved by the landmarker weka.classifiers.trees.DecisionStump
2240.5
Maximum of means among attributes of the numeric type.
Minimal mutual information between the nominal attributes and the target attribute.
3.03
Second quartile (Median) of skewness among attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1
0
Number of attributes divided by the number of instances.
Maximum mutual information between the nominal attributes and the target attribute.
The minimal number of distinct values among attributes of the nominal type.
0
Percentage of binary attributes.
236.95
Second quartile (Median) of standard deviation of attributes of the numeric type.
Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1
Number of attributes needed to optimally describe the class (under the assumption of independence among attributes). Equals ClassEntropy divided by MeanMutualInformation.
The maximum number of distinct values among attributes of the nominal type.
0
Minimum skewness among attributes of the numeric type.
0
Percentage of instances having missing values.
Third quartile of entropy among attributes.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .00001
17.12
Maximum skewness among attributes of the numeric type.
0
Minimum standard deviation of attributes of the numeric type.
0
Percentage of missing values.
279.9
Third quartile of kurtosis among attributes of the numeric type.
0.99
Average class difference between consecutive instances.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2
Error rate achieved by the landmarker weka.classifiers.trees.J48 -C .00001
1293.84
Maximum standard deviation of attributes of the numeric type.
Percentage of instances belonging to the least frequent class.
100
Percentage of numeric attributes.
1545
Third quartile of means among attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.DecisionStump -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2
Kappa coefficient achieved by the landmarker weka.classifiers.trees.J48 -C .00001
Average entropy of the attributes.
Number of instances belonging to the least frequent class.
0
Percentage of nominal attributes.
Third quartile of mutual information between the nominal attributes and the target attribute.
Error rate achieved by the landmarker weka.classifiers.trees.DecisionStump -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .0001
113.13
Mean kurtosis among attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.bayes.NaiveBayes
First quartile of entropy among attributes.
12.32
Third quartile of skewness among attributes of the numeric type.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.DecisionStump -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3
Error rate achieved by the landmarker weka.classifiers.trees.J48 -C .0001
634.57
Mean of means among attributes of the numeric type.
Error rate achieved by the landmarker weka.classifiers.bayes.NaiveBayes
-1.2
First quartile of kurtosis among attributes of the numeric type.
892.29
Third quartile of standard deviation of attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.bayes.NaiveBayes -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3
Kappa coefficient achieved by the landmarker weka.classifiers.trees.J48 -C .0001
Average mutual information between the nominal attributes and the target attribute.
Kappa coefficient achieved by the landmarker weka.classifiers.bayes.NaiveBayes
0.01
First quartile of means among attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 1
Error rate achieved by the landmarker weka.classifiers.bayes.NaiveBayes -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .001
An estimate of the amount of irrelevant information in the attributes regarding the class. Equals (MeanAttributeEntropy - MeanMutualInformation) divided by MeanMutualInformation.
0
Number of binary attributes.
First quartile of mutual information between the nominal attributes and the target attribute.
Error rate achieved by the landmarker weka.classifiers.trees.REPTree -L 1
Kappa coefficient achieved by the landmarker weka.classifiers.bayes.NaiveBayes -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Area Under the ROC Curve achieved by the landmarker weka.classifiers.lazy.IBk -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Standard deviation of the number of distinct values among attributes of the nominal type.
Error rate achieved by the landmarker weka.classifiers.trees.J48 -C .001
Average number of distinct values among the attributes of the nominal type.
0
First quartile of skewness among attributes of the numeric type.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 1
Error rate achieved by the landmarker weka.classifiers.lazy.IBk -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Area Under the ROC Curve achieved by the landmarker weka.classifiers.lazy.IBk
Kappa coefficient achieved by the landmarker weka.classifiers.trees.J48 -C .001
5.53
Mean skewness among attributes of the numeric type.
0.06
First quartile of standard deviation of attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 2
Kappa coefficient achieved by the landmarker weka.classifiers.lazy.IBk -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.BestFirst -D 1 -N 5" -W
Error rate achieved by the landmarker weka.classifiers.lazy.IBk
Percentage of instances belonging to the most frequent class.
404.33
Mean standard deviation of attributes of the numeric type.
Second quartile (Median) of entropy among attributes.
Error rate achieved by the landmarker weka.classifiers.trees.REPTree -L 2
Entropy of the target attribute values.
Kappa coefficient achieved by the landmarker weka.classifiers.lazy.IBk
Number of instances belonging to the most frequent class.
Minimal entropy among attributes.
8.27
Second quartile (Median) of kurtosis among attributes of the numeric type.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 2
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 3
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.DecisionStump
Maximum entropy among attributes.
-1.2
Minimum kurtosis among attributes of the numeric type.
82.83
Second quartile (Median) of means among attributes of the numeric type.
Error rate achieved by the landmarker weka.classifiers.trees.REPTree -L 3
Error rate achieved by the landmarker weka.classifiers.trees.DecisionStump
505.36
Maximum kurtosis among attributes of the numeric type.
0
Minimum of means among attributes of the numeric type.
Second quartile (Median) of mutual information between the nominal attributes and the target attribute.

11 tasks

0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
Define a new task