Projet

Général

Profil

Documentation » Historique » Version 36

François Rioult, 19/10/2013 20:46

1 3 François Rioult
h1. Documentation
2 1 François Rioult
3 4 François Rioult
KDAriane is a set of operators for data mining and machine learning, and a set of scenarios (supervised classification, missing values completion, strong emerging pattern mining, etc.). It uses Ariane as a graphical platform for designing the data streams.
4
5 30 François Rioult
h2. Demonstration
6
7
You can watch the [[short demonstration of KDAriane]].
8
9 1 François Rioult
h2. Installation
10 4 François Rioult
11 13 François Rioult
* [[Prerequisite]]
12 4 François Rioult
* [[KDAriane]] 
13 1 François Rioult
14 25 François Rioult
h2. Tutorial
15
16 26 François Rioult
You can read the document [[Data preparation]] for a first contact with @KDAriane@.
17 25 François Rioult
18 28 François Rioult
h2. Scenarios
19
20
In the @sample@ folder, @KDAriane@ provides some examples of KDD scenarios:
21
* [[Data preparation]]: a first scenario for the binarization of CSV data.
22
* [[Data mining complexity]]: pattern mining and complexity visualization
23
* [[CMAR]]: supervised classification with association rules [1]
24
25 17 François Rioult
h2. How is Ariane working?
26 1 François Rioult
27 17 François Rioult
Ariane is a graphical platform for designing image processing streams. Ariane works with graphical operators, and allows to build loops and while.
28
"Go to the Aiane dedicated site":http://www.greyc.ensicaen.fr/~regis/Ariane/ for a full documentation about Ariane.
29 18 François Rioult
30 6 François Rioult
KDAriane is provided with basic components for executing shell scripts. When an operator is executed, Ariane launches the script with giving the following arguments:
31 17 François Rioult
<pre>
32 33 François Rioult
script.sh parameter-1 ... parameter-p input-1 ... input-i output-1 ... output-o
33 24 François Rioult
</pre>
34
35 29 François Rioult
See [[Eval operators]] for shell scripting or calling binaries.
36 24 François Rioult
37 5 François Rioult
h2. Operators
38 1 François Rioult
39 28 François Rioult
The operator are divided in three categories: 
40
41 29 François Rioult
* [[Eval operators]] that directly execute the commands entered by Ariane. 
42 1 François Rioult
* [[KDD operators]] are special components for calling Weka components or RapidMiner processes.
43 10 François Rioult
* [[Shell operators]] that directly execute the commands entered by Ariane. 
44
45 27 François Rioult
h2. Pattern mining prototypes
46 1 François Rioult
47 28 François Rioult
KDAriane uses some specific prototypes: 
48
49 27 François Rioult
* [[music-dfs]] : mining patterns under various constraints
50
* [[mtminer]] : levelwise minimal transversals of hypergraph
51 28 François Rioult
* [[mvminer]] : mining delta-free patterns in presence of missing values
52 1 François Rioult
53
h2. Routines
54
55
Routines are complex streams that Ariane allows you to share.
56
* [[shell routines]]
57 31 François Rioult
* [[routines for missing values ]]
58 1 François Rioult
* [[KDD routines]]
59 31 François Rioult
60 34 François Rioult
h2. To-do
61
62
* sous-projet members
63
* unifier options mvminer
64 24 François Rioult
65
h2. Appendix
66 1 François Rioult
67 24 François Rioult
h3. Utilities
68
69 27 François Rioult
* [[gnuplot]]: plotting data with GNUplot
70 36 François Rioult
* [[score]] : computing validation values from a decision file
71 24 François Rioult
* [[complement]] :
72
* [[dictionary]] :
73
* [[segmentation]] :
74
75
h3. References