% % !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! % % Survival treated as the class attribute % % As used by Kilpatrick, D. & Cameron-Jones, M. (1998). Numeric prediction % using instance-based learning with encoding length selection. In Progress % in Connectionist-Based Information Systems. Singapore: Springer-Verlag. % % !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! % % 1. Title: Echocardiogram Data % % 2. Source Information: % -- Donor: Steven Salzberg (salzberg@cs.jhu.edu) % -- Collector: % -- Dr. Evlin Kinney % -- The Reed Institute % -- P.O. Box 402603 % -- Maimi, FL 33140-0603 % -- Date Received: 28 February 1989 % % 3. Past Usage: % -- 1. Salzberg, S. (1988). Exemplar-based learning: Theory and % implementation (Technical Report TR-10-88). Harvard University, % Center for Research in Computing Technology, Aiken Computation % Laboratory (33 Oxford Street; Cambridge, MA 02138). % -- Steve applied his EACH program to predict survival (i.e., life % or death), did not use the wall-motion attribute, and recorded 87 % correct and 29 incorrect in an incremental application to this % database. He also showed that, by tuning EACH to this domain, % EACH was able to derive (non-incrementally) a set of 28 % hyper-rectangles that could perfectly classify 119 instances. % -- 2. Kan, G., Visser, C., Kooler, J., & Dunning, A. (1986). Short % and long term predictive value of wall motion score in acute % myocardial infarction. British Heart Journal, 56, 422-427. % -- They predicted the same variable (whether patients will live % one year after a heart attack) using a different set of 345 % instances. Their statistical test recorded a 61% accuracy % in predicting that a patient will die (post-hoc fit). % -- 3. Elvin Kinney (in communication with Steven Salzberg) reported % that a Cox regression application recorded a 60% accuracy % in predicting that a patient will die. % % 4. Relevant Information: % -- All the patients suffered heart attacks at some point in the past. % Some are still alive and some are not. The survival and still-alive % variables, when taken together, indicate whether a patient survived % for at least one year following the heart attack. % % The problem addressed by past researchers was to predict from the % other variables whether or not the patient will survive at least % one year. The most difficult part of this problem is correctly % predicting that the patient will NOT survive. (Part of the difficulty % seems to be the size of the data set.) % % 5. Number of Instances: 132 % % 6. Number of Attributes: 13 (all numeric-valued) % % 7. Attribute Information: % 1. survival -- the number of months patient survived (has survived, % if patient is still alive). Because all the patients % had their heart attacks at different times, it is % possible that some patients have survived less than % one year but they are still alive. Check the second % variable to confirm this. Such patients cannot be % used for the prediction task mentioned above. % 2. still-alive -- a binary variable. 0=dead at end of survival period, % 1 means still alive % 3. age-at-heart-attack -- age in years when heart attack occurred % 4. pericardial-effusion -- binary. Pericardial effusion is fluid % around the heart. 0=no fluid, 1=fluid % 5. fractional-shortening -- a measure of contracility around the heart % lower numbers are increasingly abnormal % 6. epss -- E-point septal separation, another measure of contractility. % Larger numbers are increasingly abnormal. % 7. lvdd -- left ventricular end-diastolic dimension. This is % a measure of the size of the heart at end-diastole. % Large hearts tend to be sick hearts. % 8. wall-motion-score -- a measure of how the segments of the left % ventricle are moving % 9. wall-motion-index -- equals wall-motion-score divided by number of % segments seen. Usually 12-13 segments are seen % in an echocardiogram. Use this variable INSTEAD % of the wall motion score. % 10. mult -- a derivate var which can be ignored % 11. name -- the name of the patient (I have replaced them with "name") % 12. group -- meaningless, ignore it % 13. alive-at-1 -- Boolean-valued. Derived from the first two attributes. % 0 means patient was either dead after 1 year or had % been followed for less than 1 year. 1 means patient % was alive at 1 year. % % 8. Missing Attribute Values: (denoted by "?") % Attribute #: Number of Missing Values: (total: 132) % ------------ ------------------------- % 1 2 % 2 1 % 3 5 % 4 1 % 5 8 % 6 15 % 7 11 % 8 4 % 9 1 % 10 4 % 11 0 % 12 22 % 13 58 % % 9. Distribution of attribute number 2: still-alive % Value Number of instances with this value % ---- ----------------------------------- % 0 88 (dead) % 1 43 (alive) % ? 1 % Total 132 % % % 10. Distribution of attribute number 13: alive-at-1 % Value Number of instances with this value % ---- ----------------------------------- % 0 50 % 1 24 % ? 58 % Total 132 % @relation 'echoMonths' @attribute still_alive {0,1} @attribute age numeric @attribute pericardial {0,1} @attribute fractional numeric @attribute epss numeric @attribute lvdd numeric @attribute wall_score numeric @attribute wall_index numeric @attribute alive_at_1 {0,1} @attribute class numeric @data 0,71,0,0.26,9,4.6,14.0,1.0,0.0,11 0,72,0,0.38,6,4.1,14.0,1.7,0.0,19 0,55,0,0.26,4,3.42,14.0,1.0,0.0,16 0,60,0,0.253,12,4.603,16.0,1.45,0.0,57 1,57,0,0.16,22,5.75,18.0,2.25,0.0,19 0,68,0,0.26,5,4.31,12.0,1.0,0.0,26 0,62,0,0.23,31,5.43,22.5,1.875,0.0,13 0,60,0,0.33,8,5.25,14.0,1.0,0.0,50 0,46,0,0.34,0,5.09,16.0,1.14,0.0,19 0,54,0,0.14,13,4.49,15.5,1.19,0.0,25 1,77,0,0.13,16,4.23,18.0,1.8,1.0,10 0,62,1,0.45,9,3.6,16.0,1.14,0.0,52 0,73,0,0.33,6,4.0,14.0,1.0,0.0,52 0,60,0,0.15,10,3.73,14.0,1.0,0.0,44 1,62,0,0.12,23,5.8,11.67,2.33,1.0,0 0,55,1,0.25,12,4.29,14.0,1.0,0.0,24 1,69,1,0.26,11,4.65,18.0,1.64,1.0,0 1,62,1,0.07,20,5.2,24.0,2.0,1.0,0 1,66,0,0.09,17,5.819,8.0,1.333,0.0,22 1,66,1,0.22,15,5.4,27.0,2.25,1.0,1 1,69,0,0.15,12,5.39,19.5,1.625,1.0,0 1,85,1,0.18,19,5.46,13.83,1.38,1.0,0 1,73,0,0.23,12,6.06,7.5,1.5,1.0,0 1,71,0,0.17,0,4.65,8.0,1.0,1.0,5 0,64,0,0.19,5,3.48,10.0,1.11,nan,48 0,54,0,0.3,7,3.85,10.0,1.667,nan,29 0,35,0,0.3,5,4.17,14.0,1.0,nan,29 0,55,0,nan,7,nan,2.0,1.0,nan,29 1,75,0,nan,-2147483648,nan,nan,1.0,nan,0 0,55,1,0.21,4,4.16,14.0,1.56,0.0,36 1,65,0,0.15,-2147483648,5.05,10.0,1.0,nan,1 1,52,1,0.17,17,5.32,14.0,1.17,nan,1 1,-2147483648,0,nan,12,nan,6.0,3.0,nan,3 0,47,0,0.4,5,3.1,12.0,1.0,nan,27 0,63,0,nan,10,nan,14.0,1.17,0.0,35 0,61,0,0.61,13,4.07,13.0,1.625,0.0,26 0,63,1,nan,-2147483648,5.31,5.0,1.0,0.0,16 1,65,0,0.06,23,nan,21.5,2.15,1.0,1 0,68,0,0.51,-2147483648,3.88,15.0,1.67,0.0,19 0,80,0,0.41,5,4.36,nan,1.0,nan,31 0,54,0,0.35,9,3.63,11.0,1.222,0.0,32 0,70,1,0.27,4,4.49,22.0,2.0,0.0,16 0,79,0,0.15,17,4.27,13.0,1.3,0.0,40 0,56,0,0.33,-2147483648,3.59,14.0,1.0,0.0,46 1,67,1,0.44,9,3.96,17.5,1.45,nan,2 0,64,0,0.09,-2147483648,nan,12.0,2.0,nan,37 1,81,0,0.12,-2147483648,nan,9.0,1.25,0.0,19 1,59,0,0.03,21,6.29,17.0,1.31,0.0,20 1,63,1,nan,-2147483648,nan,23.0,2.3,1.0,0 1,56,1,0.04,14,5.0,nan,nan,1.0,2 1,61,1,0.27,-2147483648,nan,9.0,1.5,1.0,7 0,57,0,0.24,14,5.26,18.0,1.38,nan,10 0,58,0,0.3,9,3.49,14.0,1.0,0.0,12 1,60,0,0.01,24,5.65,39.0,3.0,1.0,1 0,66,0,0.29,15,6.15,14.0,1.0,0.0,10 0,63,0,0.15,13,4.57,13.0,1.08,0.0,45 0,57,0,0.13,18,4.37,12.33,1.37,0.0,22 0,70,0,0.1,9,5.3,23.0,2.3,0.0,53 0,68,0,0.29,-2147483648,4.41,14.0,1.167,nan,38 0,79,0,0.17,11,5.15,10.5,1.05,0.0,26 0,73,0,0.12,-2147483648,6.78,16.67,1.39,nan,9 0,72,0,0.187,12,5.02,13.0,1.18,0.0,26 1,59,0,0.13,16,4.96,17.83,1.37,nan,0 0,67,1,0.11,10,4.68,11.0,1.0,nan,12 0,51,0,0.16,13,5.26,11.0,1.0,0.0,49 1,50,0,0.14,11,4.75,10.0,2.5,nan,0 0,70,1,0.25,9,5.57,5.5,1.1,0.0,49 0,65,0,0.36,8,5.78,12.0,1.0,0.0,47 0,78,0,0.06,16,5.62,13.67,1.367,0.0,41 1,86,0,0.225,12,5.2,24.0,2.18,1.0,0 0,56,0,0.25,11,4.72,11.0,1.0,0.0,33 0,60,0,0.12,10,4.31,15.0,1.67,0.0,29 0,59,0,0.29,7,4.75,13.0,1.08,0.0,41 0,50,0,0.06,30,5.95,21.5,2.39,nan,26 0,54,0,0.217,17,4.54,16.5,1.18,0.0,15 1,68,0,0.22,21,4.85,15.0,1.15,nan,0 1,-2147483648,0,0.26,19,4.77,21.0,2.1,1.0,0 0,64,0,0.2,7,4.58,14.0,1.0,0.0,12 0,63,0,0.2,5,5.2,8.0,1.0,nan,32 0,65,0,0.06,23,6.74,12.0,1.09,nan,32 0,54,1,0.07,16,4.16,18.0,1.5,0.0,27 0,62,0,0.25,6,4.48,11.0,1.0,nan,23 1,78,0,0.05,10,4.44,15.0,1.36,1.0,0 1,61,0,nan,-2147483648,nan,28.0,2.33,1.0,0 0,52,0,0.14,25,6.21,11.5,1.15,nan,34 1,73,0,0.05,14,4.14,15.5,1.41,nan,1 1,70,1,0.16,19,5.25,11.0,1.0,nan,21 0,55,0,0.28,5,4.48,22.0,1.83,0.0,55 1,60,0,0.18,8,4.56,13.5,1.04,nan,15 1,67,0,0.155,11,5.16,13.0,1.0,nan,0 0,64,0,0.3,6,4.36,14.0,1.27,nan,35 0,59,0,0.344,9,4.04,9.0,1.0,0.0,53 0,46,0,0.272,16,5.36,12.67,1.06,nan,33 0,63,0,0.25,5,3.87,18.0,1.5,nan,33 1,74,0,0.2,4,4.56,12.5,1.04,0.0,40 0,59,0,0.5,9,3.42,18.0,1.5,nan,33 1,65,1,0.16,8,5.47,16.0,1.45,1.0,5 1,58,0,0.17,28,6.73,26.08,2.01,1.0,4 0,53,0,0.17,-2147483648,4.69,10.0,1.0,nan,31 0,66,0,0.2,-2147483648,4.23,12.0,1.0,0.0,33 0,70,0,0.38,0,4.55,10.0,1.0,0.0,22 0,62,0,0.258,11,4.87,11.0,1.0,nan,25 1,63,0,0.3,6,3.52,18.16,1.51,1.0,1 0,59,0,0.17,14,5.49,13.5,1.5,0.0,24 0,57,0,0.228,9,4.29,11.0,1.0,0.0,25 0,57,0,0.036,7,4.12,13.5,1.23,nan,24 1,78,0,0.23,40,6.23,14.0,1.4,1.0,0 1,62,0,0.26,7,4.42,14.0,1.0,1.0,3 0,62,0,0.22,12,3.92,11.0,1.0,nan,27 0,66,0,0.24,13,4.38,22.0,2.2,nan,13 0,61,0,0.27,9,4.06,12.0,1.0,nan,36 0,59,1,0.4,9,5.36,12.0,1.0,nan,25 0,57,0,0.29,9,4.77,9.0,1.0,nan,27 0,62,1,0.19,28,6.63,19.5,1.95,nan,34 0,-2147483648,0,0.26,0,4.38,9.0,1.0,nan,37 0,54,0,0.43,9,4.79,10.0,1.0,nan,34 1,62,1,0.24,28,5.86,21.5,1.95,nan,28 0,-2147483648,0,0.23,19,5.49,12.0,1.2,nan,28 0,64,0,0.15,6,4.17,14.0,1.27,nan,17 0,57,1,0.12,0,2.32,16.5,1.375,nan,38 0,61,0,0.18,0,4.48,11.0,1.375,nan,31 0,61,1,0.19,13,5.04,19.0,1.73,nan,12 0,48,0,0.15,12,3.66,10.0,1.0,nan,36 0,-2147483648,0,0.09,6,4.96,13.0,1.08,nan,17 0,61,0,0.14,25,5.16,14.0,1.27,nan,21 1,64,0,0.24,12,4.72,12.0,1.0,nan,7 0,64,0,0.28,5,5.47,11.0,1.1,nan,41 0,69,0,0.2,7,5.05,14.5,1.21,nan,36 0,57,0,0.14,16,4.36,15.0,1.36,nan,22 0,62,0,0.15,0,4.51,15.5,1.409,nan,20