Data
ada_prior

ada_prior

active ARFF Publicly available Visibility: public Uploaded 06-10-2014 by Joaquin Vanschoren
0 likes downloaded by 8 people , 8 total downloads 0 issues 0 downvotes
  • derived invalid_ARFF mythbusting_1 study_144 study_15 study_20 study_41 study_52
Issue #Downvotes for this reason By


Loading wiki
Help us complete this description Edit
Author: Source: Unknown - Date unknown Please cite: Datasets from the Agnostic Learning vs. Prior Knowledge Challenge (http://www.agnostic.inf.ethz.ch) Dataset from: http://www.agnostic.inf.ethz.ch/datasets.php Modified by TunedIT (converted to ARFF format) ADA is the marketing database The task of ADA is to discover high revenue people from census data. This is a two-class classification problem. The raw data from the census bureau is known as the Adult database in the UCI machine-learning repository. The 14 original attributes (features) include age, workclass, education, education, marital status, occupation, native country, etc. It contains continuous, binary and categorical features. This dataset is from "prior knowledge track", i.e. has access to the original features and their identity. Number of examples: Pos_ex Neg_ex Tot_ex Train 1029 3118 4147 Valid 103 312 415 This dataset contains samples from both training and validation datasets. ### Attribute information 1. age Instance’s age (numeric) 2. workclass Instance’s work class (nominal) 3. fnlwgt Instance’s sampling weight (numeric) 4. education Instance’s education level (nominal) 5. educationNum Instance’s education level (numeric version) 6. maritalStatus Instance’s marital status (nominal) 7. occupation Instance’s occupation (nominal) 8. relationship Instance’s type of relationship (nominal) 9. race Instance’s race (nominal) 10. sex Instance’s sex (nominal) 11. capitalGain Instance’s capital gain (numeric) 12. capitalLoss Instance’s capital loss (numeric) 13. hoursPerWeek Instance’s number of working hours (numeric) 14. nativeCountry Instance’s native country (numeric) 15. label Class attribute (1: the instance earns more than 50K a year; -1 otherwise)

15 features

label (target)nominal2 unique values
0 missing
agenumeric70 unique values
0 missing
workclassnominal7 unique values
0 missing
fnlwgtnumeric4222 unique values
0 missing
educationnominal16 unique values
0 missing
educationNumnumeric16 unique values
0 missing
maritalStatusnominal7 unique values
0 missing
occupationnominal14 unique values
0 missing
relationshipnominal6 unique values
0 missing
racenominal5 unique values
0 missing
sexnominal2 unique values
0 missing
capitalGainnumeric72 unique values
0 missing
capitalLossnumeric57 unique values
0 missing
hoursPerWeeknumeric77 unique values
0 missing
nativeCountrynominal39 unique values
88 missing

19 properties

4562
Number of instances (rows) of the dataset.
15
Number of attributes (columns) of the dataset.
2
Number of distinct values of the target attribute (if it is nominal).
88
Number of missing values in the dataset.
88
Number of instances with at least one value missing.
6
Number of numeric attributes.
9
Number of nominal attributes.
1.93
Percentage of instances having missing values.
0.62
Average class difference between consecutive instances.
0.13
Percentage of missing values.
0
Number of attributes divided by the number of instances.
40
Percentage of numeric attributes.
75.19
Percentage of instances belonging to the most frequent class.
60
Percentage of nominal attributes.
3430
Number of instances belonging to the most frequent class.
24.81
Percentage of instances belonging to the least frequent class.
1132
Number of instances belonging to the least frequent class.
2
Number of binary attributes.
13.33
Percentage of binary attributes.

16 tasks

520 runs - estimation_procedure: 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: label
189 runs - estimation_procedure: 10 times 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: label
0 runs - estimation_procedure: 33% Holdout set - target_feature: label
69 runs - estimation_procedure: 10-fold Learning Curve - target_feature: label
0 runs - estimation_procedure: Interleaved Test then Train - target_feature: label
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