Citation Request:
   This primary tumor domain was obtained from the University Medical Centre,
   Institute of Oncology, Ljubljana, Yugoslavia.  Thanks go to M. Zwitter and 
   M. Soklic for providing the data.  Please include this citation if you plan
   to use this database.

1. Title: Primary Tumor Domain

2. Sources:
     (a) Source:
     (b) Donors: Igor Kononenko, 
                 University E.Kardelj
                 Faculty for electrical engineering
                 Trzaska 25
                 61000 Ljubljana (tel.: (38)(+61) 265-161

                 Bojan Cestnik
                 Jozef Stefan Institute
                 Jamova 39
                 61000 Ljubljana
                 Yugoslavia (tel.: (38)(+61) 214-399 ext.287) 
     (c) Date: November 1988

3. Past Usage: (sveral)
    1. Cestnik,G., Konenenko,I, & Bratko,I. (1987). Assistant-86: A
       Knowledge-Elicitation Tool for Sophisticated Users.  In I.Bratko
       & N.Lavrac (Eds.) Progress in Machine Learning, 31-45, Sigma Press.
       -- Assistant-86: 44% accuracy
    2. Clark,P. & Niblett,T. (1987). Induction in Noisy Domains.  In
       I.Bratko & N.Lavrac (Eds.) Progress in Machine Learning, 11-30,
       Sigma Press.
       -- Simple Bayes: 48% accuracy
       -- CN2 (95% threshold): 45%
    3. Michalski,R., Mozetic,I. Hong,J., & Lavrac,N. (1986).  The Multi-Purpose
       Incremental Learning System AQ15 and its Testing Applications to Three
       Medical Domains.  In Proceedings of the Fifth National Conference on
       Artificial Intelligence, 1041-1045. Philadelphia, PA: Morgan Kaufmann.
       -- Experts: 42% accuracy 
       -- AQ15: 29-41%

4. Relevant Information:
     This is one of three domains provided by the Oncology Institute
     that has repeatedly appeared in the machine learning literature.
     (See also breast-cancer and lymphography.)

5. Number of Instances: 339

6. Number of Attributes: 18 including the class attribute

7. Attribute Information: (class is location of tumor)
    --- NOTE: All attribute values in the database have been entered as
              numeric values corresponding to their index in the list
              of attribute values for that attribute domain as given below.
    1. class: lung, head & neck, esophasus, thyroid, stomach, duoden & sm.int,
              colon, rectum, anus, salivary glands, pancreas, gallblader,
              liver, kidney, bladder, testis, prostate, ovary, corpus uteri, 
              cervix uteri, vagina, breast
    2. age:   <30, 30-59, >=60
    3. sex:   male, female
    4. histologic-type: epidermoid, adeno, anaplastic
    5. degree-of-diffe: well, fairly, poorly
    6. bone: yes, no
    7. bone-marrow: yes, no
    8. lung: yes, no
    9. pleura: yes, no
   10. peritoneum: yes, no
   11. liver: yes, no
   12. brain: yes, no
   13. skin: yes, no
   14. neck: yes, no
   15. supraclavicular: yes, no
   16. axillar: yes, no
   17. mediastinum: yes, no
   18. abdominal: yes, no

8. Missing Attribute Values: (? indicates unknown value)
    Attribute#: Number of missing values
    1: 0
    2: 0
    3: 1
    4: 67
    5: 155
    6: 0
    7: 0
    8: 0
    9: 0
    10: 0
    11: 0
    12: 0
    13: 1
    14: 0
    15: 0
    16: 1
    17: 0
    18: 0

9. Class Distribution: 
    Class Index:   Number of instances in class:
              1:   84
              2:   20
              3:   9
              4:   14
              5:   39
              6:   1
              7:   14
              8:   6
              9:   0
	     10:   2
	     11:   28
	     12:   16
	     13:   7
	     14:   24
	     15:   2
	     16:   1
	     17:   10
	     18:   29
	     19:   6
	     20:   2
	     21:   1
	     22:   24