Author:
Source: Unknown -
Please cite:
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
Relabeled values in attribute age
From: 1 To: '<30'
From: 2 To: '30-59'
From: 3 To: '>=60'
Relabeled values in attribute sex
From: 1 To: male
From: 2 To: female
Relabeled values in attribute histologic-type
From: 1 To: epidermoid
From: 2 To: adeno
From: 3 To: anaplastic
Relabeled values in attribute degree-of-diffe
From: 1 To: well
From: 2 To: fairly
From: 3 To: poorly
Relabeled values in attribute bone
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute bone-marrow
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute lung
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute pleura
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute peritoneum
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute liver
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute brain
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute skin
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute neck
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute supraclavicular
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute axillar
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute mediastinum
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute abdominal
From: 1 To: yes
From: 2 To: no
Relabeled values in attribute class
From: 1 To: lung
From: 2 To: 'head and neck'
From: 3 To: esophagus
From: 4 To: thyroid
From: 5 To: stomach
From: 6 To: 'duoden and sm.int'
From: 7 To: colon
From: 8 To: rectum
From: 9 To: anus
From: 10 To: 'salivary glands'
From: 11 To: pancreas
From: 12 To: gallbladder
From: 13 To: liver
From: 14 To: kidney
From: 15 To: bladder
From: 16 To: testis
From: 17 To: prostate
From: 18 To: ovary
From: 19 To: 'corpus uteri'
From: 20 To: 'cervix uteri'
From: 21 To: vagina
From: 22 To: breast
-----
We have redefined the number of classes to account for the real number of observations.