Blame view

help/dataminingHelp 3.39 KB
16035364   Benjamin Renard   First commit
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
<h2><img src="js/resources/images/16x16/search.png" style="margin: 0px 10px 0px 0px;"/>Data Mining</h2> 
<p>
DataMining module is searching in data for time intervals when a condition on parameters (defined by user)
is valid.<br/><br/>
<b>1 : </b>To construct a condition <b><i>drag</i></b> the parameter[s] from the Parameter Tree and <b><i>drop</i></b>  
it onto the <b><i>Data Mining Condition</i></b> area.<br/>
<img class="centered" src="help/images/mining1.png"/><br/>
</p>
<p>
<b>2 : </b>While constructing you may use different components from the <b><i>Tools For Condition Construction</i></b> -
for example  - <b><i>Constants</i></b>:<br/><br/> 
<img class="centered" src="help/images/derived2.png"><br/> 
</p>
<p> 
<b>3 : </b>To combine and compare different AMDA parameters which possibly have different 
sampling times you have to specify the  condition <b><i>Sampling Time Step</i></b>.<br/>
</p><p> 
<img align="left" src="help/images/mining2.png" style="margin: 0px 10px 0px 0px;"/>
Depending on their original sampling, AMDA parameters will be averaged or interpolated to obtain 
a condition evaluated at a <b><i>Sampling Time Step</i></b> resolution. Before evaluate the expression
all the parameters are resampled and after that the expression is calculated. 
 </p>
<p>
You may also put  <b><i>Data Gap</i></b> value. It is expressed in the parameter 'sampling 
times' and defines  gaps threshold. If data is absent for time interval greater than this threshold
the interval is treated as 'gap', otherwise the interval is filled with the interpolated data.
</p>
<p>
Parameter' <i>Sampling Time</i> as well as <i>Global Start - Stop Time Interval</i>
are displayed in popup window shown when the mouse is over the parameter name in Parameters tree.
</p><br/>
<p>
<img align="left"  src="help/images/mining3.png" style="margin: 0px 10px 0px 0px;"/>
<b>4 : </b>Data Mining could be done either for a single time interval 
(<b><i>Start Time</i></b> and <b><i>Time Interval</i></b> should be defined) 
496ef6c8   Myriam Bouchemit   modif help
36
or for a set of time intervals from one or several <b><i>Time Tables or Catalogs</i></b>.
16035364   Benjamin Renard   First commit
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
</p><br/>
<p>
<b>5 : </b>Condition  may be saved in your Workspace for next use. 
Just give the name to a condition (<b><i>Request Name</i></b>) and 
click the <b><i>Save Request</i></b> button. Your condition is added to the <b><i>DataMining</i></b>
tree of the Workspace Explorer, <b><i>Operations</i></b> tab.
<img class="centered" src="help/images/mining4.png"/><br/>
</p>
<p>
<img align="left" src="help/images/mining6.png" style="margin: 0px 10px 0px 0px;"/><br/>
<b>6 : </b>Double clicking on condition name in the <b><i>DataMining</i></b> tree 
(Workspace Explorer, <b><i>Operations</i></b> tab) or selecting <b><i>DataMining By Request</i></b>
from the context menu opens the DataMining module filled with the condition content.
</p>
<br/><br/><br/><br/><br/><br/><br/><br/><br/><br/>
<p>
<b>7 : </b> Data Mining produces two XML tables (Time Table  and  Gaps Time Table) accessible from the <b><i>Results</i></b> window : 
<img class="centered" src="help/images/mining8.png"><br/><br/>
</p>
<p>
<img align="right" src="help/images/mining9.png" style="margin: 0px 0px 0px 10px;"/> 
<b>8 : </b> <b><i>Results</i></b> window may be always opened by selecting <b><i>Show Result</i></b> from context menu opened on the right click at the corresponding node of the
 FinishedJobs / Results, DataMining tree of the Workspace Explorer, <b><i>Jobs</i></b> tab.
496ef6c8   Myriam Bouchemit   modif help
60
</p>