Data
thyroid-dis

thyroid-dis

active ARFF Publicly available Visibility: public Uploaded 26-07-2016 by Rafael G. Mantovani
1 likes downloaded by 8 people , 16 total downloads 0 issues 0 downvotes
Issue #Downvotes for this reason By


Loading wiki
Help us complete this description Edit
General Description of Thyroid Disease Databases and Related Files This directory contains 6 databases, corresponding test set, and corresponding documentation. They were left at the University of California at Irvine by Ross Quinlan during his visit in 1987 for the 1987 Machine Learning Workshop. The documentation files (with file extension "names") are formatted to be read by Quinlan's C4 decision tree program. Though briefer than the other documentation files found in this database repository, they should suffice to describe the database, specifically: 1. Source 2. Number and names of attributes (including class names) 3. Types of values that each attribute takes In general, these databases are quite similar and can be characterized somewhat as follows: 1. Many attributes (29 or so, mostly the same set over all the databases) 2. mostly numeric or Boolean valued attributes 3. thyroid disease domains (records provided by the Garavan Institute of Sydney, Australia) 4. several missing attribute values (signified by "?") 5. small number of classes (under 10, changes with each database) 7. 2800 instances in each data set 8. 972 instances in each test set (It seems that the test sets' instances are disjoint with respect to the corresponding data sets, but this has not been verified) See the following for a discussion of relevant experiments and related work: Quinlan,J.R., Compton,P.J., Horn,K.A., & Lazurus,L. (1986). Inductive knowledge acquisition: A case study. In Proceedings of the Second Australian Conference on Applications of Expert Systems. Sydney, Australia. Quinlan,J.R. (1986). Induction of decision trees. Machine Learning, 1, 81--106. Note that the instances in these databases are followed by a vertical bar and a number. These appear to be a patient id number. The vertical bar is interpreted by Quinlan's algorithms as "ignore the remainder of this line". ====================================================================== This database now also contains an additional two data files, named hypothyroid.data and sick-euthyroid.data. They have approximately the same data format and set of attributes as the other 6 databases, but their integrity is questionable. Ross Quinlan is concerned that they may have been corrupted since they first arrived at UCI, but we have not yet established the validity of this possibility. These 2 databases differ in terms of their number of instances (3163) and lack of corresponding test files. They each have 2 concepts (negative/hypothyroid and sick-euthyroid/negative respectively). Their source also appears to be the Garavan institute. Each contains several missing values. Another relatively recent file thyroid0387.data has been added that contains the latest version of an archive of thyroid diagnoses obtained from the Garvan Institute, consisting of 9172 records from 1984 to early 1987. A domain theory related to thyroid disease has also been added recently (thyroid.theory). The files new-thyroid.[names,data] were donated by Stefan Aberhard.

27 features

Class (target)nominal5 unique values
0 missing
V1numeric94 unique values
0 missing
V2nominal2 unique values
0 missing
V3nominal2 unique values
0 missing
V4nominal2 unique values
0 missing
V5nominal2 unique values
0 missing
V6nominal2 unique values
0 missing
V7nominal2 unique values
0 missing
V8nominal2 unique values
0 missing
V9nominal2 unique values
0 missing
V10nominal2 unique values
0 missing
V11nominal2 unique values
0 missing
V12nominal2 unique values
0 missing
V13nominal2 unique values
0 missing
V14nominal2 unique values
0 missing
V15nominal2 unique values
0 missing
V16nominal2 unique values
0 missing
V17nominal2 unique values
0 missing
V18numeric264 unique values
0 missing
V19nominal2 unique values
0 missing
V20numeric65 unique values
0 missing
V21nominal2 unique values
0 missing
V22numeric218 unique values
0 missing
V23nominal2 unique values
0 missing
V24numeric139 unique values
0 missing
V25nominal2 unique values
0 missing
V26numeric210 unique values
0 missing

62 properties

2800
Number of instances (rows) of the dataset.
27
Number of attributes (columns) of the dataset.
5
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.
6
Number of numeric attributes.
21
Number of nominal attributes.
1.36
First quartile of skewness among attributes of the numeric type.
46.57
Mean of means among attributes of the numeric type.
0.6
First quartile of standard deviation of attributes of the numeric type.
0.43
Average class difference between consecutive instances.
0.03
Average mutual information between the nominal attributes and the target attribute.
0.26
Second quartile (Median) of entropy among attributes.
1.53
Entropy of the target attribute values.
8.47
An estimate of the amount of irrelevant information in the attributes regarding the class. Equals (MeanAttributeEntropy - MeanMutualInformation) divided by MeanMutualInformation.
11.74
Second quartile (Median) of kurtosis among attributes of the numeric type.
0.01
Number of attributes divided by the number of instances.
2.14
Average number of distinct values among the attributes of the nominal type.
28.26
Second quartile (Median) of means among attributes of the numeric type.
48.56
Number of attributes needed to optimally describe the class (under the assumption of independence among attributes). Equals ClassEntropy divided by MeanMutualInformation.
4.05
Mean skewness among attributes of the numeric type.
0.02
Second quartile (Median) of mutual information between the nominal attributes and the target attribute.
58.29
Percentage of instances belonging to the most frequent class.
17.84
Mean standard deviation of attributes of the numeric type.
1.73
Second quartile (Median) of skewness among attributes of the numeric type.
1632
Number of instances belonging to the most frequent class.
0
Minimal entropy among attributes.
74.07
Percentage of binary attributes.
20.39
Second quartile (Median) of standard deviation of attributes of the numeric type.
0.89
Maximum entropy among attributes.
4.62
Minimum kurtosis among attributes of the numeric type.
0
Percentage of instances having missing values.
0.48
Third quartile of entropy among attributes.
317.57
Maximum kurtosis among attributes of the numeric type.
1
Minimum of means among attributes of the numeric type.
0
Percentage of missing values.
118.84
Third quartile of kurtosis among attributes of the numeric type.
110.79
Maximum of means among attributes of the numeric type.
0
Minimal mutual information between the nominal attributes and the target attribute.
22.22
Percentage of numeric attributes.
109.5
Third quartile of means among attributes of the numeric type.
0.12
Maximum mutual information between the nominal attributes and the target attribute.
2
The minimal number of distinct values among attributes of the nominal type.
77.78
Percentage of nominal attributes.
0.04
Third quartile of mutual information between the nominal attributes and the target attribute.
5
The maximum number of distinct values among attributes of the nominal type.
1.32
Minimum skewness among attributes of the numeric type.
0.11
First quartile of entropy among attributes.
5.81
Third quartile of skewness among attributes of the numeric type.
15.65
Maximum skewness among attributes of the numeric type.
0.18
Minimum standard deviation of attributes of the numeric type.
7.02
First quartile of kurtosis among attributes of the numeric type.
31.88
Third quartile of standard deviation of attributes of the numeric type.
34.21
Maximum standard deviation of attributes of the numeric type.
1.11
Percentage of instances belonging to the least frequent class.
31
Number of instances belonging to the least frequent class.
1.77
First quartile of means among attributes of the numeric type.
0.65
Standard deviation of the number of distinct values among attributes of the nominal type.
0.3
Average entropy of the attributes.
20
Number of binary attributes.
0
First quartile of mutual information between the nominal attributes and the target attribute.
67.68
Mean kurtosis among attributes of the numeric type.

5 tasks

31 runs - estimation_procedure: 10-fold Crossvalidation - target_feature: Class
0 runs - estimation_procedure: 33% Holdout set - target_feature: Class
0 runs - estimation_procedure: 50 times Clustering
0 runs - estimation_procedure: 50 times Clustering
Define a new task