Data
vowel

vowel

active ARFF Publicly available Visibility: public Uploaded 22-08-2014 by Joaquin Vanschoren
0 likes downloaded by 13 people , 17 total downloads 0 issues 0 downvotes
  • OpenML-CC18 OpenML100 study_1 study_123 study_135 study_14 study_144 study_34 study_50 study_52 study_7 study_76 study_98 study_99
Issue #Downvotes for this reason By


Loading wiki
Help us complete this description Edit
Author: Peter Turney (peter@ai.iit.nrc.ca) Source: [UCI](https://archive.ics.uci.edu/ml/machine-learning-databases/undocumented/connectionist-bench/vowel/) - date unknown Please cite: [UCI citation policy](https://archive.ics.uci.edu/ml/citation_policy.html) Vowel Recognition (Deterding data) Speaker independent recognition of the eleven steady state vowels of British English using a specified training set of lpc derived log area ratios. Collected by David Deterding (data and non-connectionist analysis), Mahesan Niranjan (first connectionist analysis), Tony Robinson (description, program, data, and results) A very comprehensive description including comments by the authors can be found [here](https://archive.ics.uci.edu/ml/machine-learning-databases/undocumented/connectionist-bench/vowel/vowel.names) The problem is specified by the accompanying data file, "vowel.data". This consists of a three dimensional array: voweldata [speaker, vowel, input]. The speakers are indexed by integers 0-89. (Actually, there are fifteen individual speakers, each saying each vowel six times.) The vowels are indexed by integers 0-10. For each utterance, there are ten floating-point input values, with array indices 0-9. The problem is to train the network as well as possible using only on data from "speakers" 0-47, and then to test the network on speakers 48-89, reporting the number of correct classifications in the test set. For a more detailed explanation of the problem, see the excerpt from Tony Robinson's Ph.D. thesis in the COMMENTS section. In Robinson's opinion, connectionist problems fall into two classes, the possible and the impossible. He is interested in the latter, by which he means problems that have no exact solution. Thus the problem here is not to see how fast a network can be trained (although this is important), but to maximise a less than perfect performance. #### METHODOLOGY Report the number of test vowels classified correctly, (i.e. the number of occurences when distance of the correct output to the actual output was the smallest of the set of distances from the actual output to all possible target outputs). Though this is not the focus of Robinson's study, it would also be useful to report how long the training took (measured in pattern presentations or with a rough count of floating-point operations required) and what level of success was achieved on the training and testing data after various amounts of training. Of course, the network topology and algorithm used should be precisely described as well. #### VARIATIONS This benchmark is proposed to encourage the exploration of different node types. Please theorise/experiment/hack. The author (Robinson) will try to correspond by email if requested. In particular there has been some discussion recently on the use of a cross-entropy distance measure, and it would be interesting to see results for that. #### Notes 1. Each of these numbers is based on a single trial with random starting weights. More trials would of course be preferable, but the computational facilities available to Robinson were limited. 2. Graphs are given in Robinson's thesis showing test-set performance vs. epoch count for some of the training runs. In most cases, performance peaks at around 250 correct, after which performance decays to different degrees. The numbers given above are final performance figures after about 3000 trials, not the peak performance obtained during the run. #### REFERENCES [Deterding89] D. H. Deterding, 1989, University of Cambridge, "Speaker Normalisation for Automatic Speech Recognition", submitted for PhD. [NiranjanFallside88] M. Niranjan and F. Fallside, 1988, Cambridge University Engineering Department, "Neural Networks and Radial Basis Functions in Classifying Static Speech Patterns", CUED/F-INFENG/TR.22. [RenalsRohwer89-ijcnn] Steve Renals and Richard Rohwer, "Phoneme Classification Experiments Using Radial Basis Functions", Submitted to the International Joint Conference on Neural Networks, Washington, 1989. [RabinerSchafer78] L. R. Rabiner and R. W. Schafer, Englewood Cliffs, New Jersey, 1978, Prentice Hall, "Digital Processing of Speech Signals". [PragerFallside88] R. W. Prager and F. Fallside, 1988, Cambridge University Engineering Department, "The Modified Kanerva Model for Automatic Speech Recognition", CUED/F-INFENG/TR.6. [BroomheadLowe88] D. Broomhead and D. Lowe, 1988, Royal Signals and Radar Establishment, Malvern, "Multi-variable Interpolation and Adaptive Networks", RSRE memo, #4148. [RobinsonNiranjanFallside88-tr] A. J. Robinson and M. Niranjan and F. Fallside, 1988, Cambridge University Engineering Department, "Generalising the Nodes of the Error Propagation Network", CUED/F-INFENG/TR.25. [Robinson89] A. J. Robinson, 1989, Cambridge University Engineering Department, "Dynamic Error Propagation Networks". [McCullochAinsworth88] N. McCulloch and W. A. Ainsworth, Proceedings of Speech'88, Edinburgh, 1988, "Speaker Independent Vowel Recognition using a Multi-Layer Perceptron". [RobinsonFallside88-neuro] A. J. Robinson and F. Fallside, 1988, Proceedings of nEuro'88, Paris, June, "A Dynamic Connectionist Model for Phoneme Recognition. #### Notes * This is version 2. Version 1 is hidden because it includes a feature dividing the data in train and test set. In OpenML this information is explicitly available in the corresponding task.

13 features

Class (target)nominal11 unique values
0 missing
Speaker_Numbernominal15 unique values
0 missing
Sexnominal2 unique values
0 missing
Feature_0numeric853 unique values
0 missing
Feature_1numeric877 unique values
0 missing
Feature_2numeric815 unique values
0 missing
Feature_3numeric836 unique values
0 missing
Feature_4numeric803 unique values
0 missing
Feature_5numeric798 unique values
0 missing
Feature_6numeric748 unique values
0 missing
Feature_7numeric794 unique values
0 missing
Feature_8numeric788 unique values
0 missing
Feature_9numeric775 unique values
0 missing

107 properties

990
Number of instances (rows) of the dataset.
13
Number of attributes (columns) of the dataset.
11
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.
10
Number of numeric attributes.
3
Number of nominal attributes.
An estimate of the amount of irrelevant information in the attributes regarding the class. Equals (MeanAttributeEntropy - MeanMutualInformation) divided by MeanMutualInformation.
1
Number of binary attributes.
0
First quartile of mutual information between the nominal attributes and the target attribute.
0.64
Error rate achieved by the landmarker weka.classifiers.trees.REPTree -L 1
0.66
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
0.68
Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3
0.87
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .001
9.33
Average number of distinct values among the attributes of the nominal type.
-0.01
First quartile of skewness among attributes of the numeric type.
0.3
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 1
0.86
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
6.66
Standard deviation of the number of distinct values among attributes of the nominal type.
0.32
Error rate achieved by the landmarker weka.classifiers.trees.J48 -C .001
0.09
Mean skewness among attributes of the numeric type.
0.57
First quartile of standard deviation of attributes of the numeric type.
0.84
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 2
0.31
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
0.96
Area Under the ROC Curve achieved by the landmarker weka.classifiers.lazy.IBk
0.65
Kappa coefficient achieved by the landmarker weka.classifiers.trees.J48 -C .001
0.7
Mean standard deviation of attributes of the numeric type.
2.45
Second quartile (Median) of entropy among attributes.
0.64
Error rate achieved by the landmarker weka.classifiers.trees.REPTree -L 2
0.66
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
0.08
Error rate achieved by the landmarker weka.classifiers.lazy.IBk
9.09
Percentage of instances belonging to the most frequent class.
1
Minimal entropy among attributes.
-0.43
Second quartile (Median) of kurtosis among attributes of the numeric type.
0.3
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 2
3.46
Entropy of the target attribute values.
0.91
Kappa coefficient achieved by the landmarker weka.classifiers.lazy.IBk
90
Number of instances belonging to the most frequent class.
-0.76
Minimum kurtosis among attributes of the numeric type.
-0.04
Second quartile (Median) of means among attributes of the numeric type.
0.84
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 3
0.68
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.DecisionStump
3.91
Maximum entropy among attributes.
-3.2
Minimum of means among attributes of the numeric type.
0
Second quartile (Median) of mutual information between the nominal attributes and the target attribute.
0.64
Error rate achieved by the landmarker weka.classifiers.trees.REPTree -L 3
0.82
Error rate achieved by the landmarker weka.classifiers.trees.DecisionStump
0.15
Maximum kurtosis among attributes of the numeric type.
0
Minimal mutual information between the nominal attributes and the target attribute.
0.06
Second quartile (Median) of skewness among attributes of the numeric type.
0.3
Kappa coefficient achieved by the landmarker weka.classifiers.trees.REPTree -L 3
0.09
Kappa coefficient achieved by the landmarker weka.classifiers.trees.DecisionStump
1.88
Maximum of means among attributes of the numeric type.
2
The minimal number of distinct values among attributes of the nominal type.
7.69
Percentage of binary attributes.
0.63
Second quartile (Median) of standard deviation of attributes of the numeric type.
0.85
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1
0.01
Number of attributes divided by the number of instances.
0
Maximum mutual information between the nominal attributes and the target attribute.
-0.21
Minimum skewness among attributes of the numeric type.
0
Percentage of instances having missing values.
3.91
Third quartile of entropy among attributes.
0.29
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.
15
The maximum number of distinct values among attributes of the nominal type.
0.46
Minimum standard deviation of attributes of the numeric type.
0
Percentage of missing values.
-0.24
Third quartile of kurtosis among attributes of the numeric type.
0
Average class difference between consecutive instances.
0.68
Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 1
0.87
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .00001
0.36
Maximum skewness among attributes of the numeric type.
9.09
Percentage of instances belonging to the least frequent class.
76.92
Percentage of numeric attributes.
0.54
Third quartile of means among attributes of the numeric type.
0.86
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
0.85
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2
0.32
Error rate achieved by the landmarker weka.classifiers.trees.J48 -C .00001
1.18
Maximum standard deviation of attributes of the numeric type.
90
Number of instances belonging to the least frequent class.
23.08
Percentage of nominal attributes.
0
Third quartile of mutual information between the nominal attributes and the target attribute.
0.31
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
0.29
Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2
0.65
Kappa coefficient achieved by the landmarker weka.classifiers.trees.J48 -C .00001
2.45
Average entropy of the attributes.
0.94
Area Under the ROC Curve achieved by the landmarker weka.classifiers.bayes.NaiveBayes
1
First quartile of entropy among attributes.
0.25
Third quartile of skewness among attributes of the numeric type.
0.66
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
0.68
Kappa coefficient achieved by the landmarker weka.classifiers.trees.RandomTree -depth 2
0.87
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.J48 -C .0001
-0.39
Mean kurtosis among attributes of the numeric type.
0.42
Error rate achieved by the landmarker weka.classifiers.bayes.NaiveBayes
-0.56
First quartile of kurtosis among attributes of the numeric type.
0.79
Third quartile of standard deviation of attributes of the numeric type.
0.86
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
0.85
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3
0.32
Error rate achieved by the landmarker weka.classifiers.trees.J48 -C .0001
-0.1
Mean of means among attributes of the numeric type.
0
Average mutual information between the nominal attributes and the target attribute.
0.54
Kappa coefficient achieved by the landmarker weka.classifiers.bayes.NaiveBayes
-0.36
First quartile of means among attributes of the numeric type.
0.84
Area Under the ROC Curve achieved by the landmarker weka.classifiers.trees.REPTree -L 1
0.31
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
0.29
Error rate achieved by the landmarker weka.classifiers.trees.RandomTree -depth 3
0.65
Kappa coefficient achieved by the landmarker weka.classifiers.trees.J48 -C .0001

34 tasks

13903 runs - estimation_procedure: 10-fold Crossvalidation - target_feature: Class
0 runs - estimation_procedure: 10 times 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: Class
0 runs - estimation_procedure: 33% Holdout set - evaluation_measure: predictive_accuracy - target_feature: Class
0 runs - estimation_procedure: 10-fold Crossvalidation - target_feature: Class
0 runs - estimation_procedure: 33% Holdout set - target_feature: Class
44 runs - estimation_procedure: 10-fold Learning Curve - target_feature: Class
0 runs - estimation_procedure: Interleaved Test then Train - target_feature: Class
0 runs - target_feature: Class
1319 runs - target_feature: Class
1317 runs - target_feature: Class
1315 runs - target_feature: Class
1315 runs - target_feature: Class
1312 runs - target_feature: Class
1308 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
0 runs - target_feature: Class
Define a new task