source: misc/publications/2017/journals/ASOC/README.txt @ 14794

Last change on this file since 14794 was 14794, checked in by gkronber, 5 years ago

#2757 updated readme

File size: 3.4 KB
Line 
11. Title: Miba Friction Plate Testing Data
22. Sources:
3   (a) Miba frictec, Andreas Promberger, Peter Mitterbauer Str. 1,
4       A-4661 Roitham, AUSTRIA, +43 76139020, Andreas.promberger@miba.com
5   (b) FH Upper Austria, Gabriel Kronberger, Softwarepark 11,
6       A-4232 Hagenberg, AUSTRIA, +43 50804 22320,
7       gabriel.kronberger@heuristiclab.com
8   (c) April 2017
9
10
113. Past Usage:
12  - G. Kronberger, M. Kommenda, E. Lughofer, S. Saminger-Platz,
13    A. Promberger, F. Nickel, S. Winkler, M. Affenzeller - Robust
14    Generalized Fuzzy Modeling and Enhanced Symbolic Regression for
15    Modeling Tribological Systems, submitted to Applied Soft
16    Computing, 2017
17
18  - E. Lughofer, G. Kronberger, M. Kommenda, S. Saminger-Platz,
19    A. Promberger, F. Nickel, S. M. Winkler, M. Affenzeller - Robust
20    Fuzzy Modeling and Symbolic Regression for Establishing Accurate
21    and Interpretable Prediction Models in Supervising Tribological
22    Systems - Proceedings of the 8th International Joint Conference on
23    Computational Intelligence, Porto, Portugal, 2016, pp. 51-63
24
254. Relevant information:
26  A set of datasets for regression modelling of
27  friction characteristics of friction plate systems. The data stem
28  from tests of friction plates with commercial test benches for wet
29  friction plate systems. Friction characteristics such as the
30  coefficient of friction, wear, and temperatures are measured at
31  different loads. The goal is to predict these values given load
32  parameters. Data have been kindly provided by Miba frictec company.
33
345. Number of instances
35  A separate file is provided for each target variable. The values of
36  the target variable are given in the last column.
37
38  - Cf1: 815 instances
39  - Cf2: 2921 instances
40  - Cf3: 657 instances
41  - Cf4: 649 instances
42  - NvhRating: 3943 instances
43  - Temp1: 656 instances
44  - Temp2: 648 instances
45  - Wear1: 904 instances
46  - Wear2: 902 instances
47
486. Number of Attributes: 28
49  (2 binary, 22 numeric & continuous, and four nominal) Depending on
50  the target variable (or file) some of the attributes might be
51  constant.
52
537. Attribute Information
54  The first column 'Partition' contains the partition assignment that
55  should be used for validation of algorithms. This attribute should
56  not be used for modelling. Rows must not be shuffled as in standard
57  cross-validation.
58
59  - Source1:      indicates that the row stems from data source one [binary]
60  - Source2:      indicates that the row stems from data source two (
61                  redundant given Source1) [binary]
62  - x1:           an attribute of the testing procedure [numeric, integer]
63  - Material_Cat: represents the type of friction material [nominal]
64  - x2, ... ,x16: material attributes  [numeric, continuous]
65  - Material:     represents the friction material [nominal]
66  - Grooving:     represents the grooving (surface structure) on the
67                  friction plate [nominal]
68  - Oil:          represents the oil type [nominal]
69  - x17, ... ,x22 load attributes [numeric, continuous]
70
71
72If you use these data files please use the following reference:
73  G. Kronberger, M. Kommenda, E. Lughofer, S. Saminger-Platz,
74  A. Promberger, F. Nickel, S. Winkler, M. Affenzeller - Robust
75  Generalized Fuzzy Modeling and Enhanced Symbolic Regression for
76  Modeling Tribological Systems, submitted to Applied Soft Computing,
77  2017
78
Note: See TracBrowser for help on using the repository browser.