SPECTF (Q6034168)

**Author**: Krzysztof J. Cios","Lukasz A. \N**Source**: [original](https://archive.ics.uci.edu/ml/datasets/SPECTF+Heart) - Date unknown \N**Please cite**: \N\NThis 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). \NThis file fixes that problem by merging the training set with the correct test set, resulting 267 instances.\N\NSPECTF heart data:\N\NThis 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.\N\NNOTE: See the SPECT heart data for binary data for the same classification task.\N\NSources: -- 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\N\NRelevant 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).