Data
meta

meta

active ARFF Publicly available Visibility: public Uploaded 03-10-2014 by Joaquin Vanschoren
0 likes downloaded by 2 people , 2 total downloads 0 issues 0 downvotes
Issue #Downvotes for this reason By


Loading wiki
Help us complete this description Edit
Author: Source: Unknown - Date unknown Please cite: 1. Title: meta-data 2. Sources: (a) Creator: LIACC - University of Porto R.Campo Alegre 823 4150 PORTO (b) Donor: P.B.Brazdil or J.Gama Tel.: +351 600 1672 LIACC, University of Porto Fax.: +351 600 3654 Rua Campo Alegre 823 Email: statlog-adm@ncc.up.pt 4150 Porto, Portugal (c) Date: March, 1996 (d) Acknowlegements: LIACC wishes to thank Commission of European Communities for their support. Also, we wish to thank the following partners for providing the individual test results: - Dept. of Statistics, University of Strathclyde, Glasgow, UK - Dept. of Statistics, University of Leeds, UK - Aston University, Birmingham, UK - Forschungszentrum Ulm, Daimler-Benz AG, Germany - Brainware GmbH, Berlin, Germany - Frauenhofer Gesellschaft IITB-EPO, Berlin, Germany - Institut fuer Kybernetik, Bochum, Germany - ISoft, Gif sur Yvette, France - Dept. of CS and AI, University of Granada, Spain 3. Past Usage: Meta-Data was used in order to give advice about which classification method is appropriate for a particular dataset. This work is described in: -"Machine Learning, Neural and Statistical Learning" Eds. D.Michie,D.J.Spiegelhalter and C.Taylor Ellis Horwood-1994 - "Characterizing the Applicability of Classification Algorithms Using Meta-Level Learning", P. Brazdil, J.Gama and B.Henery: in Proc. of Machine Learning - ECML-94, ed. F.Bergadano and L.de Raedt,LNAI Vol.784 Springer-Verlag. -"Characterization of Classification Algorithms" J.Gama, P.Brazdil in Proc. of EPIA 95, LNAI Vol.990 Springer-Verlag, 1995 4. Relevant Information:n This DataSet is about the results of Statlog project. The project performed a comparative study between Statistical, Neural and Symbolic learning algorithms. Project StatLog (Esprit Project 5170) was concerned with comparative studies of different machine learning, neural and statistical classification algorithms. About 20 different algorithms were evaluated on more than 20 different datasets. The tests carried out under project produced many interesting results. Algorithms DataSets ------------------------- -------------------------- C4.5 NewId Credit_Austr Belgian AC2 CART Chromosome Credit_Man IndCART Cal5 CUT DNA CN2 ITRule Diabetes Digits44 Discrim QuaDisc Credit_German Faults LogDisc ALLOC80 Head Heart kNN SMART KLDigits Letters BayesTree CASTLE New_Belgian Sat_Image DIPLO92 RBF Segment Shuttle LVQ Backprop Technical TseTse Kohonen Vehicle The results of these tests are comprehensively described in a book (D.Michie et.al, 1994). 5. Number of Instances: 528 6. Number of Attributes: 22 (including an Id#) plus the class attribute -- all but two attributes are continuously valued 7. Attribute Information: 1. DS_Name categorical Name of DataSet 2. T continuous Number of examples in test set 3. N continuous Number of examples 4. p continuous Number of attributes 5. k continuous Number of classes 6. Bin continuous Number of binary Attributes 7. Cost continuous Cost (1=yes,0=no) 8. SDratio continuous Standard deviation ratio 9. correl continuous Mean correlation between attributes 10. cancor1 continuous First canonical correlation 11. cancor2 continuous Second canonical correlation 12. fract1 continuous First eigenvalue 13. fract2 continuous Second eigenvalue 14. skewness continuous Mean of |E(X-Mean)|^3/STD^3 15. kurtosis continuous Mean of |E(X-Mean)|^4/STD^4 16. Hc continuous Mean entropy of attributes 17. Hx continuous Entropy of classes 18. MCx continuous Mean mutual entropy of class and attributes 19. EnAtr continuous Equivalent number of attributes 20. NSRatio continuous Noise-signal ratio 21. Alg_Name categorical Name of Algorithm 22. Norm_error continuous Normalized Error (continuous class) 8. Missing Attribute Values: Note that fract2 and cancor2 only apply to datasets with more than 2 classes. When they appear as '?' this means a don't care value. Summary Statistics: Attribute Min Max Mean Std T 270 20000 4569.05 5704.01 N 270 58000 10734.2 14568.8 p 6 180 29.5455 36.8533 k 2 91 9.72727 19.3568 Bin 0 43 3.18182 9.29227 Cost 0 1 0.13636 0.35125 SdRatio 1.0273 4.0014 1.4791 0.65827 Correl 0.0456 0.751 0.23684 0.1861 Cancor1 0.5044 0.9884 0.79484 0.15639 Cancor2 0.1057 0.9623 0.74106 0.269 Fract1 0.1505 1 0.70067 0.3454 Fract2 0.2807 1 0.70004 0.29405 Skewness 0.1802 6.7156 1.78422 1.79022 Kurtosis 0.9866 160.311 22.6672 41.8496 Hc 0.2893 4.8787 1.87158 1.44665 Hx 0.3672 6.5452 3.34502 1.80383 Mcx 0.0187 1.3149 0.31681 0.33548 EnAtr 1.56006 160.644 20.6641 35.6614 NsRatio 1.02314 159.644 28.873 37.925

22 features

class (target)numeric436 unique values
0 missing
DS_Namenominal22 unique values
0 missing
Tnumeric20 unique values
0 missing
Nnumeric20 unique values
0 missing
pnumeric18 unique values
0 missing
knumeric9 unique values
0 missing
Binnumeric7 unique values
0 missing
Costnumeric2 unique values
0 missing
SDrationumeric22 unique values
0 missing
correlnumeric21 unique values
24 missing
cancor1numeric22 unique values
0 missing
cancor2numeric12 unique values
240 missing
fract1numeric13 unique values
0 missing
fract2numeric10 unique values
240 missing
skewnessnumeric22 unique values
0 missing
kurtosisnumeric22 unique values
0 missing
Hcnumeric21 unique values
0 missing
Hxnumeric22 unique values
0 missing
MCxnumeric22 unique values
0 missing
EnAtrnumeric22 unique values
0 missing
NSRationumeric22 unique values
0 missing
Alg_Namenominal24 unique values
0 missing

107 properties

528
Number of instances (rows) of the dataset.
22
Number of attributes (columns) of the dataset.
0
Number of distinct values of the target attribute (if it is nominal).
504
Number of missing values in the dataset.
264
Number of instances with at least one value missing.
20
Number of numeric attributes.
2
Number of nominal attributes.
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
14247.3
Maximum standard deviation of attributes of the numeric type.
Percentage of instances belonging to the least frequent class.
90.91
Percentage of numeric attributes.
27.32
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.
9.09
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
14.87
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.
2.96
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
776.48
Mean of means among attributes of the numeric type.
Error rate achieved by the landmarker weka.classifiers.bayes.NaiveBayes
-0.04
First quartile of kurtosis among attributes of the numeric type.
36.83
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
0.71
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
1.41
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
23
Average number of distinct values among the attributes of the nominal type.
0.27
First quartile of skewness among attributes of the numeric type.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 1
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
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
2.2
Mean skewness among attributes of the numeric type.
0.33
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
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
Error rate achieved by the landmarker weka.classifiers.lazy.IBk
Percentage of instances belonging to the most frequent class.
1038.72
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
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
Kappa coefficient achieved by the landmarker weka.classifiers.lazy.IBk
Number of instances belonging to the most frequent class.
Minimal entropy among attributes.
2.24
Second quartile (Median) of kurtosis among attributes of the numeric type.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 2
Entropy of the target attribute values.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.DecisionStump
Maximum entropy among attributes.
-1.58
Minimum kurtosis among attributes of the numeric type.
2.53
Second quartile (Median) of means among attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 3
Error rate achieved by the landmarker weka.classifiers.trees.DecisionStump
225.98
Maximum kurtosis among attributes of the numeric type.
0.14
Minimum of means among attributes of the numeric type.
Second quartile (Median) of mutual information between the nominal attributes and the target attribute.
Error rate achieved by the landmarker weka.classifiers.trees.REPTree -L 3
Kappa coefficient achieved by the landmarker weka.classifiers.trees.DecisionStump
10734.18
Maximum of means among attributes of the numeric type.
Minimal mutual information between the nominal attributes and the target attribute.
1.85
Second quartile (Median) of skewness among attributes of the numeric type.
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 3
0.04
Number of attributes divided by the number of instances.
Maximum mutual information between the nominal attributes and the target attribute.
22
The minimal number of distinct values among attributes of the nominal type.
0
Percentage of binary attributes.
1.76
Second quartile (Median) of standard deviation of attributes of the numeric type.
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1
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.
24
The maximum number of distinct values among attributes of the nominal type.
-1.51
Minimum skewness among attributes of the numeric type.
50
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
14.69
Maximum skewness among attributes of the numeric type.
0.15
Minimum standard deviation of attributes of the numeric type.
4.34
Percentage of missing values.
8.86
Third quartile of kurtosis among attributes of the numeric type.
-172.88
Average class difference between consecutive instances.

7 tasks

32 runs - estimation_procedure: 10-fold Crossvalidation - evaluation_measure: mean_absolute_error - target_feature: class
0 runs - estimation_procedure: 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: class
0 runs - estimation_procedure: 10 times 10-fold Crossvalidation - evaluation_measure: mean_absolute_error - target_feature: class
0 runs - estimation_procedure: 10 times 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: class
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
Define a new task