source: orange/orange/doc/datasets/balance-scale.htm @ 1760:9d4bb141fb0e

Revision 1760:9d4bb141fb0e, 2.3 KB checked in by blaz <blaz.zupan@…>, 9 years ago (diff)

data info file

Line 
1<html>
2<head>
3<title>Balance Scale Weight & Distance Data Base</title>
4</head>
5<body>
6<h1>Info on Balance Scale Weight & Distance Data Base</h1>
7<pre>
81. Title: Balance Scale Weight & Distance Database
9
102. Source Information:
11    (a) Source: Generated to model psychological experiments reported
12        by Siegler, R. S. (1976).  Three Aspects of Cognitive
13        Development.  Cognitive Psychology, 8, 481-520.
14    (b) Donor: Tim Hume (hume@ics.uci.edu)
15    (c) Date: 22 April 1994
16
173. Past Usage: (possibly different formats of this data)
18   - Publications
19    1. Klahr, D., & Siegler, R.S. (1978).  The Representation of
20       Children's Knowledge.  In H. W. Reese & L. P. Lipsitt (Eds.),
21       Advances in Child Development and Behavior, pp. 61-116.  New
22       York: Academic Press
23    2. Langley,P. (1987).  A General Theory of Discrimination
24       Learning.  In D. Klahr, P. Langley, & R. Neches (Eds.),
25       Production System Models of Learning and Development, pp.
26       99-161. Cambridge, MA: MIT Press
27    3. Newell, A. (1990).  Unified Theories of Cognition.
28       Cambridge, MA: Harvard University Press
29    4. McClelland, J.L. (1988).  Parallel Distibuted Processing:
30       Implications for Cognition and Development.  Technical
31       Report AIP-47, Department of Psychology, Carnegie-Mellon
32       University
33    5. Shultz, T., Mareschal, D., & Schmidt, W. (1994).  Modeling
34       Cognitive Development on Balance Scale Phenomena. Machine
35       Learning, Vol. 16, pp. 59-88.
36
374. Relevant Information:
38    This data set was generated to model psychological
39    experimental results.  Each example is classified as having the
40    balance scale tip to the right, tip to the left, or be
41    balanced.  The attributes are the left weight, the left
42    distance, the right weight, and the right distance.  The
43    correct way to find the class is the greater of
44    (left-distance * left-weight) and (right-distance *
45    right-weight).  If they are equal, it is balanced.
46
475. Number of Instances: 625 (49 balanced, 288 left, 288 right)
48
496. Number of Attributes: 4 (numeric) + class name = 5
50
517. Attribute Information:
52    1. Class Name: 3 (L, B, R)
53    2. Left-Weight: 5 (1, 2, 3, 4, 5)
54    3. Left-Distance: 5 (1, 2, 3, 4, 5)
55    4. Right-Weight: 5 (1, 2, 3, 4, 5)
56    5. Right-Distance: 5 (1, 2, 3, 4, 5)
57
588. Missing Attribute Values:
59    none
60
619. Class Distribution:
62   1. 46.08 percent are L
63   2. 07.84 percent are B
64   3. 46.08 percent are R
65<pre>
66</body>
67</html>
Note: See TracBrowser for help on using the repository browser.