OpenML
churn

churn

active ARFF public Visibility: public Uploaded 06-04-2017 by Pieter Gijsbers
2 likes downloaded by 9 people , 11 total downloads 0 issues 0 downvotes
  • Computer Systems Machine Learning OpenML-CC18 study_135 study_144 study_98 study_99 study_293 study_270 study_271 study_253 study_258 study_285 study_275
Issue #Downvotes for this reason By


Loading wiki
Help us complete this description Edit
Author: Unknown Source: [PMLB](https://github.com/EpistasisLab/penn-ml-benchmarks/tree/master/datasets/classification), [BigML](https://bigml.com/user/francisco/gallery/dataset/5163ad540c0b5e5b22000383), Supposedly from UCI but I can't find it there. Please cite: A dataset relating characteristics of telephony account features and usage and whether or not the customer churned. Originally used in [Discovering Knowledge in Data: An Introduction to Data Mining](http://secs.ac.in/wp-content/CSE_PORTAL/DataMining_Daniel.pdf).

21 features

class (target)nominal2 unique values
0 missing
statenumeric51 unique values
0 missing
account_lengthnumeric218 unique values
0 missing
area_codenominal3 unique values
0 missing
phone_numbernumeric5000 unique values
0 missing
international_plannominal2 unique values
0 missing
voice_mail_plannominal2 unique values
0 missing
number_vmail_messagesnumeric48 unique values
0 missing
total_day_minutesnumeric1961 unique values
0 missing
total_day_callsnumeric123 unique values
0 missing
total_day_chargenumeric1961 unique values
0 missing
total_eve_minutesnumeric1879 unique values
0 missing
total_eve_callsnumeric126 unique values
0 missing
total_eve_chargenumeric1659 unique values
0 missing
total_night_minutesnumeric1853 unique values
0 missing
total_night_callsnumeric131 unique values
0 missing
total_night_chargenumeric1028 unique values
0 missing
total_intl_minutesnumeric170 unique values
0 missing
total_intl_callsnumeric21 unique values
0 missing
total_intl_chargenumeric170 unique values
0 missing
number_customer_service_callsnominal10 unique values
0 missing
Show all 21 features

62 properties

NumberOfInstances
5000
Number of instances (rows) of the dataset.
NumberOfFeatures
21
Number of attributes (columns) of the dataset.
NumberOfClasses
2
Number of distinct values of the target attribute (if it is nominal).
NumberOfMissingValues
0
Number of missing values in the dataset.
NumberOfInstancesWithMissingValues
0
Number of instances with at least one value missing.
NumberOfNumericFeatures
16
Number of numeric attributes.
NumberOfSymbolicFeatures
5
Number of nominal attributes.
MinSkewnessOfNumericAtts
-0.21
Minimum skewness among attributes of the numeric type.
PercentageOfSymbolicFeatures
23.81
Percentage of nominal attributes.
Quartile3MutualInformation
0.05
Third quartile of mutual information between the nominal attributes and the target attribute.
MaxNominalAttDistinctValues
10
The maximum number of distinct values among attributes of the nominal type.
MinStdDevOfNumericAtts
0.75
Minimum standard deviation of attributes of the numeric type.
Quartile1AttributeEntropy
0.55
First quartile of entropy among attributes.
Quartile3SkewnessOfNumericAtts
0.02
Third quartile of skewness among attributes of the numeric type.
MaxSkewnessOfNumericAtts
1.36
Maximum skewness among attributes of the numeric type.
MinorityClassPercentage
14.14
Percentage of instances belonging to the least frequent class.
Quartile1KurtosisOfNumericAtts
-0.02
First quartile of kurtosis among attributes of the numeric type.
Quartile3StdDevOfNumericAtts
47.82
Third quartile of standard deviation of attributes of the numeric type.
MaxStdDevOfNumericAtts
1443.52
Maximum standard deviation of attributes of the numeric type.
MinorityClassSize
707
Number of instances belonging to the least frequent class.
Quartile1MeansOfNumericAtts
9.33
First quartile of means among attributes of the numeric type.
StdvNominalAttDistinctValues
3.49
Standard deviation of the number of distinct values among attributes of the nominal type.
MeanAttributeEntropy
1.27
Average entropy of the attributes.
NumberOfBinaryFeatures
3
Number of binary attributes.
Quartile1MutualInformation
0
First quartile of mutual information between the nominal attributes and the target attribute.
MeanKurtosisOfNumericAtts
0.18
Mean kurtosis among attributes of the numeric type.
Quartile1SkewnessOfNumericAtts
-0.05
First quartile of skewness among attributes of the numeric type.
MeanMeansOfNumericAtts
224.32
Mean of means among attributes of the numeric type.
Quartile1StdDevOfNumericAtts
3.15
First quartile of standard deviation of attributes of the numeric type.
AutoCorrelation
0.76
Average class difference between consecutive instances.
MeanMutualInformation
0.02
Average mutual information between the nominal attributes and the target attribute.
MeanNoiseToSignalRatio
51.46
An estimate of the amount of irrelevant information in the attributes regarding the class. Equals (MeanAttributeEntropy - MeanMutualInformation) divided by MeanMutualInformation.
Quartile2AttributeEntropy
1.17
Second quartile (Median) of entropy among attributes.
ClassEntropy
0.59
Entropy of the target attribute values.
MeanNominalAttDistinctValues
3.8
Average number of distinct values among the attributes of the nominal type.
Quartile2KurtosisOfNumericAtts
0.08
Second quartile (Median) of kurtosis among attributes of the numeric type.
Dimensionality
0
Number of attributes divided by the number of instances.
MeanSkewnessOfNumericAtts
0.14
Mean skewness among attributes of the numeric type.
Quartile2MeansOfNumericAtts
65.28
Second quartile (Median) of means among attributes of the numeric type.
EquivalentNumberOfAtts
24.36
Number of attributes needed to optimally describe the class (under the assumption of independence among attributes). Equals ClassEntropy divided by MeanMutualInformation.
MeanStdDevOfNumericAtts
109.24
Mean standard deviation of attributes of the numeric type.
Quartile2MutualInformation
0.02
Second quartile (Median) of mutual information between the nominal attributes and the target attribute.
MajorityClassPercentage
85.86
Percentage of instances belonging to the most frequent class.
MinAttributeEntropy
0.45
Minimal entropy among attributes.
Quartile2SkewnessOfNumericAtts
-0.01
Second quartile (Median) of skewness among attributes of the numeric type.
MajorityClassSize
4293
Number of instances belonging to the most frequent class.
MinKurtosisOfNumericAtts
-1.2
Minimum kurtosis among attributes of the numeric type.
PercentageOfBinaryFeatures
14.29
Percentage of binary attributes.
Quartile2StdDevOfNumericAtts
17.31
Second quartile (Median) of standard deviation of attributes of the numeric type.
MaxAttributeEntropy
2.28
Maximum entropy among attributes.
MinMeansOfNumericAtts
2.77
Minimum of means among attributes of the numeric type.
PercentageOfInstancesWithMissingValues
0
Percentage of instances having missing values.
Quartile3AttributeEntropy
2.08
Third quartile of entropy among attributes.
MaxKurtosisOfNumericAtts
3.27
Maximum kurtosis among attributes of the numeric type.
MinMutualInformation
0
Minimal mutual information between the nominal attributes and the target attribute.
PercentageOfMissingValues
0
Percentage of missing values.
Quartile3KurtosisOfNumericAtts
0.19
Third quartile of kurtosis among attributes of the numeric type.
MaxMeansOfNumericAtts
2499.5
Maximum of means among attributes of the numeric type.
MinNominalAttDistinctValues
2
The minimal number of distinct values among attributes of the nominal type.
PercentageOfNumericFeatures
76.19
Percentage of numeric attributes.
Quartile3MeansOfNumericAtts
160.28
Third quartile of means among attributes of the numeric type.
MaxMutualInformation
0.05
Maximum mutual information between the nominal attributes and the target attribute.
Show all 62 properties

26 tasks

Supervised Classification on churn
5464 runs - estimation_procedure: 10-fold Crossvalidation - target_feature: class
Supervised Classification on churn
2049 runs - estimation_procedure: 10-fold Crossvalidation - evaluation_measure: precision - target_feature: class
Supervised Classification on churn
0 runs - estimation_procedure: 10 times 10-fold Crossvalidation - target_feature: class
Supervised Classification on churn
0 runs - estimation_procedure: 33% Holdout set - target_feature: class
Supervised Classification on churn
0 runs - estimation_procedure: 4-fold Crossvalidation - target_feature: class
Supervised Classification on churn
0 runs - estimation_procedure: 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Learning Curve on churn
0 runs - estimation_procedure: 10-fold Learning Curve - target_feature: class
Supervised Data Stream Classification on churn
0 runs - estimation_procedure: Interleaved Test then Train - target_feature: class
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Clustering on churn
0 runs - estimation_procedure: 50 times Clustering
Define a new task