Summary

(No information yet)

License

unknown (from Weka repository)

Dependencies

Tags

arff slurped Weka

Attribute Types

Integer

Download

# Instances: 80 / # Attributes: 45
HDF5 (29.4 KB) XML CSV ARFF LibSVM Matlab Octave

Files are converted on demand and the process can take up to a minute. Please wait until download begins.

Original Data Format

arff

Name

spect

Version mldata

0

Comment

1. Title of Database: SPECTF heart data

2. 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

3. Past Usage:

   1. Kurgan, L.A., Cios, K.J., Tadeusiewicz, R., Ogiela, M. & Goodenday, L.S. "Knowledge Discovery Approach to Automated Cardiac SPECT Diagnosis" Artificial Intelligence in Medicine, vol. 23:2, pp 149-169, Oct 2001

   Results: The CLIP3 machine learning algorithm achieved 77.0% accuracy. CLIP3 references:
   Cios, K.J., Wedding, D.K. & Liu, N. CLIP3: cover learning using integer programming. Kybernetes, 26:4-5, pp 513-536, 1997

   Cios K. J. & Kurgan L. 
   Hybrid Inductive Machine Learning: An Overview of CLIP Algorithms, 
   In: Jain L.C., and Kacprzyk J. (Eds.) 
       New Learning Paradigms in Soft Computing, 
       Physica-Verlag (Springer), 2001
   

   SPECTF is a good data set for testing ML algorithms; it has 267 instances that are descibed by 45 attributes. Predicted attribute: OVERALL_DIAGNOSIS (binary) NOTE: See the SPECT heart data for binary data for the same classification task.

4. 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 cardilogists' diagnoses).

5. Number of Instances: 267

6. Number of Attributes: 45 (44 continuous + 1 binary class)

7. Attribute Information:

8. OVERALL_DIAGNOSIS: 0,1 (class attribute, binary)

9. F1R: continuous (count in ROI (region of interest) 1 in rest)

10. F1S: continuous (count in ROI 1 in stress)

11. F2R: continuous (count in ROI 2 in rest)

12. F2S: continuous (count in ROI 2 in stress)

13. F3R: continuous (count in ROI 3 in rest)

14. F3S: continuous (count in ROI 3 in stress)

15. F4R: continuous (count in ROI 4 in rest)

16. F4S: continuous (count in ROI 4 in stress)

17. F5R: continuous (count in ROI 5 in rest)

18. F5S: continuous (count in ROI 5 in stress)

19. F6R: continuous (count in ROI 6 in rest)

20. F6S: continuous (count in ROI 6 in stress)

21. F7R: continuous (count in ROI 7 in rest)

22. F7S: continuous (count in ROI 7 in stress)

23. F8R: continuous (count in ROI 8 in rest)

24. F8S: continuous (count in ROI 8 in stress)

25. F9R: continuous (count in ROI 9 in rest)

26. F9S: continuous (count in ROI 9 in stress)

27. F10R: continuous (count in ROI 10 in rest)

28. F10S: continuous (count in ROI 10 in stress)

29. F11R: continuous (count in ROI 11 in rest)

30. F11S: continuous (count in ROI 11 in stress)

31. F12R: continuous (count in ROI 12 in rest)

32. F12S: continuous (count in ROI 12 in stress)

33. F13R: continuous (count in ROI 13 in rest)

34. F13S: continuous (count in ROI 13 in stress)

35. F14R: continuous (count in ROI 14 in rest)

36. F14S: continuous (count in ROI 14 in stress)

37. F15R: continuous (count in ROI 15 in rest)

38. F15S: continuous (count in ROI 15 in stress)

39. F16R: continuous (count in ROI 16 in rest)

40. F16S: continuous (count in ROI 16 in stress)

41. F17R: continuous (count in ROI 17 in rest)

42. F17S: continuous (count in ROI 17 in stress)

43. F18R: continuous (count in ROI 18 in rest)

44. F18S: continuous (count in ROI 18 in stress)

45. F19R: continuous (count in ROI 19 in rest)

46. F19S: continuous (count in ROI 19 in stress)

47. F20R: continuous (count in ROI 20 in rest)

48. F20S: continuous (count in ROI 20 in stress)

49. F21R: continuous (count in ROI 21 in rest)

50. F21S: continuous (count in ROI 21 in stress)

51. F22R: continuous (count in ROI 22 in rest)

52. F22S: continuous (count in ROI 22 in stress) -- all continuous attributes have integer values from the 0 to 100 -- dataset is divided into: -- training data ("SPECTF.train" 80 instances) -- testing data ("SPECTF.test" 187 instances)

53. Missing Attribute Values: None

54. Class Distribution: -- entire data Class # examples 0 55 1 212 -- training dataset Class # examples 0 40 1 40 -- testing dataset Class # examples 0 15 1 172

NOTE: See the SPECT heart data for binary data for the same classification task.

Information about the dataset CLASSTYPE: nominal CLASSINDEX: first

Names

OVERALL_DIAGNOSIS,F1R,F1S,F2R,F2S,F3R,F3S,F4R,F4S,F5R,

Types

1. nominal:0,1
2. numeric
3. numeric
4. numeric
5. numeric
6. numeric
7. numeric
8. numeric
9. numeric
10. numeric

Data (first 10 data points)

OVER...	F1R	F1S	F2R	F2S	F3R	F3S	F4R	F4S	F5R	...
1	59	52	70	67	73	66	72	61	58	...
1	72	62	69	67	78	82	74	65	69	...
1	71	62	70	64	67	64	79	65	70	...
1	69	71	70	78	61	63	67	65	59	...
1	70	66	61	66	61	58	69	69	72	...
1	57	69	68	75	69	74	73	71	57	...
1	69	66	62	75	67	71	72	76	69	...
1	61	60	60	62	64	72	68	67	74	...
1	65	62	67	68	65	67	71	71	64	...
1	74	73	72	79	66	61	76	66	65	...
...	...	...	...	...	...	...	...	...	...	...

Description

A gzip'ed tar containing UCI and UCI KDD datasets (uci-20070111.tar.gz, 17,952,832 Bytes)

URLs

(No information yet)

Publications

Data Source

http://www.ics.uci.edu/~mlearn/MLRepository.html http://kdd.ics.uci.edu/

Measurement Details

Usage Scenario

revision 1

by mldata on 2011-09-14 15:57

No one has posted any comments yet. Perhaps you would like to be the first?

Leave a comment

This item was downloaded 5027 times and viewed 2440 times.

No Tasks yet on dataset uci-20070111 spectf_train