Risk_classifier/Machine_Learning_based_Classifier_desc_ug.html at Fred02-12 · Sajal010/Risk_classifier · GitHub

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
<!DOCTYPE html>
<html>
<head>
  <meta charset="utf-8">
  <meta name="viewport" content="width=device-width">
  <title>MathJax example</title>
  <script src="https://polyfill.io/v3/polyfill.min.js?features=es6"></script>
  <script id="MathJax-script" async
          src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js">
  </script>
</head>

<body>

<h2 style="text-align: center;">&nbsp;<strong>Machine Learning based Classifier</strong></h2>
<div>
<h3><strong>Usage Guide:</strong></h3>
<p><span style="color: #ff6600;">In order to use Machine Learning based Classifier on your data, you need to follow these steps:</span></p>
<ol>
<li><span style="color: #ff6600;"> In 'Decision Tree' tab, the first thing that the user will see is 'Usage Guide' after that in 'View Dataset' tab the user can specify the dataset to be used. There is  a default dataset "GermanCredit.Rda" for the first time users. The user can also add their own dataset in the format as specified in 'File input'.</span></li>
<li><span style="color: #ff6600;"> After selecting the dataset, the user can also view the dataset added in the 'View Dataset' tab plus all the variables in the dataset.</span></li>
<li><span style="color: #ff6600;"> Once the dataset is added, the user can select the type of variable selction method and number of decision trees in the 'Method Specification' tab. After selection the user can initiate the algorithm by clicking the 'calculate' button at the bottom and can view the reduced dataset in the 'Dataset after variable selection' tab once the function complete running. Note : In this app, if the user is not interested in feature selection step they can directly go to Step 2 i.e. Possible subset of decision trees.</span></li>
<li><span style="color: #ff6600;"> Once the method finishes running, decision tree will appear under the 'Classifier' tab which enables user to view tree with maximum accuracy, minimum gini index and maximum AUROC plus they can download the specified tree in pdf or png format.</span></li>
</ol>
</div>

</body>
</html>