SPECTF
OpenML dataset with id 1600
No author found.
Full work available at URL: https://api.openml.org/data/v1/download/1663773/SPECTF.arff
Upload date: 27 August 2015
Dataset Characteristics
Number of classes: 2
Number of features: 45 (numeric: 44, symbolic: 1 and in total binary: 1 )
Number of instances: 267
Number of instances with missing values: 0
Number of missing values: 0
Author: Krzysztof J. Cios","Lukasz A. Source: original - Date unknown Please cite:
This is a corrected version of the previous data file in version 1, which contained a dataset (349 instances) incorrectly merged from the original training and test sets available on UCI (there are actually 2 test sets there, one of them is incorrect). This file fixes that problem by merging the training set with the correct test set, resulting 267 instances.
SPECTF heart data:
This is a merged version of the separate train and test set which are usually distributed. On OpenML this train-test split can be found as one of the possible tasks.
NOTE: See the SPECT heart data for binary data for the same classification task.
Sources: -- Original owners: Krzysztof J. Cios, Lukasz A. Kurgan University of Colorado at Denver, Denver, CO 80217, U.S.A. Krys.Cios@cudenver.edu Lucy S. Goodenday Medical College of Ohio, OH, U.S.A. -- Donors: Lukasz A.Kurgan, Krzysztof J. Cios -- Date: 10/01/01
Relevant Information: The dataset describes diagnosing of cardiac Single Proton Emission Computed Tomography (SPECT) images. Each of the patients is classified into two categories: normal and abnormal. The database of 267 SPECT image sets (patients) was processed to extract features that summarize the original SPECT images. As a result, 44 continuous feature pattern was created for each patient. The CLIP3 algorithm was used to generate classification rules from these patterns. The CLIP3 algorithm generated rules that were 77.0% accurate (as compared with cardiologists' diagnoses).
