2 |
26 Feb 07 |
jari |
1 |
<html> |
2 |
26 Feb 07 |
jari |
2 |
<body bgcolor = "#FFFFCC"><basefont face = "Arial"> |
2 |
26 Feb 07 |
jari |
3 |
<h1>SVM: Support Vector Machines</h1> |
2 |
26 Feb 07 |
jari |
4 |
<h1>Process Initialization Dialog</h1> |
2 |
26 Feb 07 |
jari |
5 |
<h2> Parameter Information</h2> |
2 |
26 Feb 07 |
jari |
6 |
<hr size = 10> |
2 |
26 Feb 07 |
jari |
7 |
|
2 |
26 Feb 07 |
jari |
8 |
<h2> Sample Selection </h2> |
2 |
26 Feb 07 |
jari |
9 |
The sample selection option indicates whether to cluster genes or experiments. |
2 |
26 Feb 07 |
jari |
10 |
<br> |
2 |
26 Feb 07 |
jari |
11 |
|
2 |
26 Feb 07 |
jari |
12 |
<h2>Process Selection</h2> |
2 |
26 Feb 07 |
jari |
13 |
The SVM algorithm works performing two main processes, training and classification. |
2 |
26 Feb 07 |
jari |
14 |
One can elect to perform training only, classification only, or both phases of the SVM classification |
2 |
26 Feb 07 |
jari |
15 |
technique. |
2 |
26 Feb 07 |
jari |
16 |
<br><br> |
2 |
26 Feb 07 |
jari |
17 |
The 'Training Only' option results in a set of numerical weights which can stored as a file and used for classification |
2 |
26 Feb 07 |
jari |
18 |
at a later time. |
2 |
26 Feb 07 |
jari |
19 |
<br><br> |
2 |
26 Feb 07 |
jari |
20 |
The 'Classification Only' option takes a file input of weights generated from training and results in a binary classification |
2 |
26 Feb 07 |
jari |
21 |
of the elements. |
2 |
26 Feb 07 |
jari |
22 |
<br><br> |
2 |
26 Feb 07 |
jari |
23 |
The 'Training and Classification' option provides the ability to use the input set as a training set to produce weights which |
2 |
26 Feb 07 |
jari |
24 |
are immediately applied to perform the classification. |
2 |
26 Feb 07 |
jari |
25 |
<br><br> |
2 |
26 Feb 07 |
jari |
26 |
The 'One-out Iterative Validation' iteratively performs an SVM training and classification run. |
2 |
26 Feb 07 |
jari |
27 |
On each iteration one element is moved to the neutral classification and therefore will not |
2 |
26 Feb 07 |
jari |
28 |
impact the SVM training nor the classification of elements. The final classification will not |
2 |
26 Feb 07 |
jari |
29 |
be biased by an initial classification of the element. |
2 |
26 Feb 07 |
jari |
30 |
<h3>** Note that the One-out Iterative Validation will iterate a training and classification |
2 |
26 Feb 07 |
jari |
31 |
run for each element being classified. This may not be practical nor necessary for the |
2 |
26 Feb 07 |
jari |
32 |
classification of large numbers of genes. In the case of genes the impact of one gene |
2 |
26 Feb 07 |
jari |
33 |
in producing a discriminate is relatively small and the time to iteratively run for |
2 |
26 Feb 07 |
jari |
34 |
each gene could be prohibitive for large sets of genes. This feature is more practical |
2 |
26 Feb 07 |
jari |
35 |
to use when classifying experiments. |
2 |
26 Feb 07 |
jari |
36 |
|
2 |
26 Feb 07 |
jari |
37 |
<br><br> |
2 |
26 Feb 07 |
jari |
38 |
|
2 |
26 Feb 07 |
jari |
39 |
<h2> Hierarchical Clustering </h2> |
2 |
26 Feb 07 |
jari |
40 |
This check box selects whether to perform hierarchical clustering on the elements in each cluster |
2 |
26 Feb 07 |
jari |
41 |
created. |
2 |
26 Feb 07 |
jari |
42 |
<br> |
2 |
26 Feb 07 |
jari |
43 |
</basefont> |
2 |
26 Feb 07 |
jari |
44 |
</body> |
2 |
26 Feb 07 |
jari |
45 |
</html> |
2 |
26 Feb 07 |
jari |
46 |
|