Data
colic

colic

active ARFF Publicly available Visibility: public Uploaded 06-04-2014 by Jan van Rijn
0 likes downloaded by 9 people , 10 total downloads 0 issues 0 downvotes
  • mythbusting_1 study_1 study_15 study_20 study_41 uci
Issue #Downvotes for this reason By


Loading wiki
Help us complete this description Edit
Author: Mary McLeish & Matt Cecile, University of Guelph Donor: Will Taylor (taylor@pluto.arc.nasa.gov) Source: [UCI](https://archive.ics.uci.edu/ml/datasets/Horse+Colic) - 8/6/89 Please cite: Horse Colic database In this version (version 2), some features were removed. It is unclear why of how this was done.

23 features

surgical_lesion (target)nominal2 unique values
0 missing
surgerynominal2 unique values
2 missing
Agenominal2 unique values
0 missing
rectal_temperaturenumeric40 unique values
69 missing
pulsenumeric54 unique values
26 missing
respiratory_ratenumeric40 unique values
71 missing
temp_extremitiesnominal4 unique values
65 missing
peripheral_pulsenominal4 unique values
83 missing
mucous_membranesnominal6 unique values
48 missing
capillary_refill_timenominal3 unique values
38 missing
painnominal5 unique values
63 missing
peristalsisnominal4 unique values
52 missing
abdominal_distensionnominal4 unique values
65 missing
nasogastric_tubenominal3 unique values
131 missing
nasogastric_refluxnominal3 unique values
133 missing
nasogastric_reflux_PHnumeric24 unique values
299 missing
rectal_examinationnominal4 unique values
128 missing
abdomennominal5 unique values
143 missing
packed_cell_volumenumeric54 unique values
37 missing
total_proteinnumeric84 unique values
43 missing
abdominocentesis_appearancenominal3 unique values
194 missing
abdomcentesis_total_proteinnumeric44 unique values
235 missing
outcomenominal3 unique values
2 missing

19 properties

368
Number of instances (rows) of the dataset.
23
Number of attributes (columns) of the dataset.
2
Number of distinct values of the target attribute (if it is nominal).
1927
Number of missing values in the dataset.
361
Number of instances with at least one value missing.
7
Number of numeric attributes.
16
Number of nominal attributes.
13.04
Percentage of binary attributes.
98.1
Percentage of instances having missing values.
0.54
Average class difference between consecutive instances.
22.77
Percentage of missing values.
0.06
Number of attributes divided by the number of instances.
30.43
Percentage of numeric attributes.
63.04
Percentage of instances belonging to the most frequent class.
69.57
Percentage of nominal attributes.
232
Number of instances belonging to the most frequent class.
36.96
Percentage of instances belonging to the least frequent class.
136
Number of instances belonging to the least frequent class.
3
Number of binary attributes.

10 tasks

772 runs - estimation_procedure: 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: surgical_lesion
352 runs - estimation_procedure: 33% Holdout set - evaluation_measure: predictive_accuracy - target_feature: surgical_lesion
332 runs - estimation_procedure: 5 times 2-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: surgical_lesion
190 runs - estimation_procedure: 10 times 10-fold Crossvalidation - evaluation_measure: predictive_accuracy - target_feature: surgical_lesion
0 runs - estimation_procedure: Leave one out - evaluation_measure: predictive_accuracy - target_feature: surgical_lesion
149 runs - estimation_procedure: 10 times 10-fold Learning Curve - evaluation_measure: predictive_accuracy - target_feature: surgical_lesion
84 runs - estimation_procedure: 10-fold Learning Curve - evaluation_measure: predictive_accuracy - target_feature: surgical_lesion
25 runs - estimation_procedure: Interleaved Test then Train - target_feature: surgical_lesion
Define a new task