source: orange/orange/doc/datasets/primary-tumor.htm @ 1760:9d4bb141fb0e

Revision 1760:9d4bb141fb0e, 3.8 KB checked in by blaz <blaz.zupan@…>, 9 years ago (diff)

data info file

Line 
1<html>
2<head>
3<title>Primary Tumor Data Base</title>
4</head>
5<body>
6<h1>Info on Primary Tumor Data Base</h1>
7<pre>
8
9Citation Request:
10   This primary tumor domain was obtained from the University Medical Centre,
11   Institute of Oncology, Ljubljana, Yugoslavia.  Thanks go to M. Zwitter and
12   M. Soklic for providing the data.  Please include this citation if you plan
13   to use this database.
14
151. Title: Primary Tumor Domain
16
172. Sources:
18     (a) Source:
19     (b) Donors: Igor Kononenko,
20                 University E.Kardelj
21                 Faculty for electrical engineering
22                 Trzaska 25
23                 61000 Ljubljana (tel.: (38)(+61) 265-161
24
25                 Bojan Cestnik
26                 Jozef Stefan Institute
27                 Jamova 39
28                 61000 Ljubljana
29                 Yugoslavia (tel.: (38)(+61) 214-399 ext.287)
30     (c) Date: November 1988
31
323. Past Usage: (sveral)
33    1. Cestnik,G., Konenenko,I, & Bratko,I. (1987). Assistant-86: A
34       Knowledge-Elicitation Tool for Sophisticated Users.  In I.Bratko
35       & N.Lavrac (Eds.) Progress in Machine Learning, 31-45, Sigma Press.
36       -- Assistant-86: 44% accuracy
37    2. Clark,P. & Niblett,T. (1987). Induction in Noisy Domains.  In
38       I.Bratko & N.Lavrac (Eds.) Progress in Machine Learning, 11-30,
39       Sigma Press.
40       -- Simple Bayes: 48% accuracy
41       -- CN2 (95% threshold): 45%
42    3. Michalski,R., Mozetic,I. Hong,J., & Lavrac,N. (1986).  The Multi-Purpose
43       Incremental Learning System AQ15 and its Testing Applications to Three
44       Medical Domains.  In Proceedings of the Fifth National Conference on
45       Artificial Intelligence, 1041-1045. Philadelphia, PA: Morgan Kaufmann.
46       -- Experts: 42% accuracy
47       -- AQ15: 29-41%
48
494. Relevant Information:
50     This is one of three domains provided by the Oncology Institute
51     that has repeatedly appeared in the machine learning literature.
52     (See also breast-cancer and lymphography.)
53
545. Number of Instances: 339
55
566. Number of Attributes: 18 including the class attribute
57
587. Attribute Information: (class is location of tumor)
59    --- NOTE: All attribute values in the database have been entered as
60              numeric values corresponding to their index in the list
61              of attribute values for that attribute domain as given below.
62    1. class: lung, head & neck, esophasus, thyroid, stomach, duoden & sm.int,
63              colon, rectum, anus, salivary glands, pancreas, gallblader,
64              liver, kidney, bladder, testis, prostate, ovary, corpus uteri,
65              cervix uteri, vagina, breast
66    2. age:   <30, 30-59, >=60
67    3. sex:   male, female
68    4. histologic-type: epidermoid, adeno, anaplastic
69    5. degree-of-diffe: well, fairly, poorly
70    6. bone: yes, no
71    7. bone-marrow: yes, no
72    8. lung: yes, no
73    9. pleura: yes, no
74   10. peritoneum: yes, no
75   11. liver: yes, no
76   12. brain: yes, no
77   13. skin: yes, no
78   14. neck: yes, no
79   15. supraclavicular: yes, no
80   16. axillar: yes, no
81   17. mediastinum: yes, no
82   18. abdominal: yes, no
83
848. Missing Attribute Values: (? indicates unknown value)
85    Attribute#: Number of missing values
86    1: 0
87    2: 0
88    3: 1
89    4: 67
90    5: 155
91    6: 0
92    7: 0
93    8: 0
94    9: 0
95    10: 0
96    11: 0
97    12: 0
98    13: 1
99    14: 0
100    15: 0
101    16: 1
102    17: 0
103    18: 0
104
1059. Class Distribution:
106    Class Index:   Number of instances in class:
107              1:   84
108              2:   20
109              3:   9
110              4:   14
111              5:   39
112              6:   1
113              7:   14
114              8:   6
115              9:   0
116         10:   2
117         11:   28
118         12:   16
119         13:   7
120         14:   24
121         15:   2
122         16:   1
123         17:   10
124         18:   29
125         19:   6
126         20:   2
127         21:   1
128         22:   24
129</pre>
130</body>
131</html>
Note: See TracBrowser for help on using the repository browser.