Creators:

1. Hungarian Institute of Cardiology. Budapest: Andras Janosi, M.D.
2. University Hospital, Zurich, Switzerland: William Steinbrunn, M.D.
3. University Hospital, basel, Switzerland: Matthias Pfisterer, M.D.
4. V.A. Medical Center, Long Beach and Cleveland Clinic Foundation: Robert Detrano, M.D., Ph.D.

Donor:

David W. Aha (aha '@' ics.uci.edu) (714) 856-8779

Data Set Information:

This database contains 76 attributes, but all published experiments refer to using a subset of 14 of them.  In particular, the Cleveland database is the only one that has been used by ML researchers to
this date.  The "goal" field refers to the presence of heart disease in the patient.  It is integer valued from 0 (no presence) to 4. Experiments with the Cleveland database have concentrated on simply attempting to distinguish presence (values 1,2,3,4) from absence (value 0).  
 
The names and social security numbers of the patients were recently removed from the database, replaced with dummy values.

One file has been "processed", that one containing the Cleveland database.  All four unprocessed files also exist in this directory.

To see Test Costs (donated by Peter Turney), please see the folder "Costs"

Attribute Information:

only 14 attributes used:
     1. #3  (age)      
     2. #4  (sex)      
     3. #9  (cp)        
     4. #10 (trestbps)  
     5. #12 (chol)      
     6. #16 (fbs)      
     7. #19 (restecg)  
     8. #32 (thalach)  
     9. #38 (exang)    
     10. #40 (oldpeak)  
     11. #41 (slope)    
     12. #44 (ca)        
     13. #51 (thal)      
     14. #58 (num)       (the predicted attribute)

Complete attribute documentation:
     1 id: patient identification number
     2 ccf: social security number (I replaced this with a dummy value of 0)
     3 age: age in years
     4 sex: sex (1 = male; 0 = female)
     5 painloc: chest pain location (1 = substernal; 0 = otherwise)
     6 painexer (1 = provoked by exertion; 0 = otherwise)
     7 relrest (1 = relieved after rest; 0 = otherwise)
     8 pncaden (sum of 5, 6, and 7)
     9 cp: chest pain type
       -- Value 1: typical angina
       -- Value 2: atypical angina
       -- Value 3: non-anginal pain
       -- Value 4: asymptomatic
    10 trestbps: resting blood pressure (in mm Hg on admission to the hospital)
    11 htn
    12 chol: serum cholestoral in mg/dl
    13 smoke: I believe this is 1 = yes; 0 = no (is or is not a smoker)
    14 cigs (cigarettes per day)
    15 years (number of years as a smoker)
    16 fbs: (fasting blood sugar > 120 mg/dl)  (1 = true; 0 = false)
    17 dm (1 = history of diabetes; 0 = no such history)
    18 famhist: family history of coronary artery disease (1 = yes; 0 = no)
    19 restecg: resting electrocardiographic results
       -- Value 0: normal
       -- Value 1: having ST-T wave abnormality (T wave inversions and/or ST elevation or depression of > 0.05 mV)
       -- Value 2: showing probable or definite left ventricular hypertrophy by Estes' criteria
    20 ekgmo (month of exercise ECG reading)
    21 ekgday(day of exercise ECG reading)
    22 ekgyr (year of exercise ECG reading)
    23 dig (digitalis used furing exercise ECG: 1 = yes; 0 = no)
    24 prop (Beta blocker used during exercise ECG: 1 = yes; 0 = no)
    25 nitr (nitrates used during exercise ECG: 1 = yes; 0 = no)
    26 pro (calcium channel blocker used during exercise ECG: 1 = yes; 0 = no)
    27 diuretic (diuretic used used during exercise ECG: 1 = yes; 0 = no)
    28 proto: exercise protocol
         1 = Bruce    
         2 = Kottus
         3 = McHenry
         4 = fast Balke
         5 = Balke
         6 = Noughton
         7 = bike 150 kpa min/min  (Not sure if "kpa min/min" is what was written!)
         8 = bike 125 kpa min/min  
         9 = bike 100 kpa min/min
        10 = bike 75 kpa min/min
        11 = bike 50 kpa min/min
        12 = arm ergometer
    29 thaldur: duration of exercise test in minutes
    30 thaltime: time when ST measure depression was noted
    31 met: mets achieved
    32 thalach: maximum heart rate achieved
    33 thalrest: resting heart rate
    34 tpeakbps: peak exercise blood pressure (first of 2 parts)
    35 tpeakbpd: peak exercise blood pressure (second of 2 parts)
    36 dummy
    37 trestbpd: resting blood pressure
    38 exang: exercise induced angina (1 = yes; 0 = no)
    39 xhypo: (1 = yes; 0 = no)
    40 oldpeak = ST depression induced by exercise relative to rest
    41 slope: the slope of the peak exercise ST segment
       -- Value 1: upsloping
       -- Value 2: flat
       -- Value 3: downsloping
    42 rldv5: height at rest
    43 rldv5e: height at peak exercise
    44 ca: number of major vessels (0-3) colored by flourosopy
    45 restckm: irrelevant
    46 exerckm: irrelevant
    47 restef: rest raidonuclid (sp?) ejection fraction
    48 restwm: rest wall (sp?) motion abnormality
       0 = none
       1 = mild or moderate
       2 = moderate or severe
       3 = akinesis or dyskmem (sp?)
    49 exeref: exercise radinalid (sp?) ejection fraction
    50 exerwm: exercise wall (sp?) motion
    51 thal: 3 = normal; 6 = fixed defect; 7 = reversable defect
    52 thalsev: not used
    53 thalpul: not used
    54 earlobe: not used
    55 cmo: month of cardiac cath (sp?)  (perhaps "call")
    56 cday: day of cardiac cath (sp?)
    57 cyr: year of cardiac cath (sp?)
    58 num: diagnosis of heart disease (angiographic disease status)
       -- Value 0: < 50% diameter narrowing
       -- Value 1: > 50% diameter narrowing
       (in any major vessel: attributes 59 through 68 are vessels)
    59 lmt
    60 ladprox
    61 laddist
    62 diag
    63 cxmain
    64 ramus
    65 om1
    66 om2
    67 rcaprox
    68 rcadist
    69 lvx1: not used
    70 lvx2: not used
    71 lvx3: not used
    72 lvx4: not used
    73 lvf: not used
    74 cathef: not used
    75 junk: not used
    76 name: last name of patient  (I replaced this with the dummy string "name")

Relevant Papers:

Detrano, R., Janosi, A., Steinbrunn, W., Pfisterer, M., Schmid, J., Sandhu, S., Guppy, K., Lee, S., & Froelicher, V. (1989).  International application of a new probability algorithm for the diagnosis of coronary artery disease. American Journal of Cardiology, 64,304--310.
[Web link]

David W. Aha & Dennis Kibler. "Instance-based prediction of heart-disease presence with the Cleveland database."
[Web link]

Gennari, J.H., Langley, P, & Fisher, D. (1989). Models of incremental concept formation. Artificial Intelligence, 40, 11--61.
[Web link]

Papers That Cite This Data Set¹:

Zhi-Hua Zhou and Yuan Jiang. NeC4.5: Neural Ensemble based C4.5. IEEE Trans. Knowl. Data Eng, 16. 2004.  [View Context].

Remco R. Bouckaert and Eibe Frank. evaluating the Replicability of Significance Tests for Comparing