Navigate to your workspace on [https://studio.deepopinion.ai](https://studio.deepopinion.ai)

Navigate to the Cognitive Skills Builder using the hamburger menu on the top left

Banking Customer Support Ticket Classification

Click "BUILD A NEW SKILL" to create a document information extraction model
Select "Document Information Extraction"
Click this text field and type a name "Invoice Information Extraction"
Click "CREATE"

Click "BUILD A NEW SKILL" to create a document information extraction model

Click "Assign exactly one label to a piece of text, e.g. routing support tickets to teams"

Assign exactly one label to a piece of text, e.g. routing support tickets to teams

This will take you into the project to start. Now you can upload your documents. Upload 1 document for now.

Once the upload is done you can move to the next step

ID	Name	# Documents	Source	Status	Created	
4675	
2022-09-26T06:13:30.442293
	1	UPLOAD	
PROCESSING
	2022-09-26 10:13	

ID	Name	# Documents	Source	Status	Created	
4675	
2022-09-26T06:13:30.442293
	1	UPLOAD	
CREATED
	2022-09-26 10:13	

Navigate to "Label Definition" from the side menu to list the information you would like to label and extract from a document

A label will be automatically created with the name "default"

You can edit it and rename it to "Invoice#"
You can also change the label color

You can also choose to upload the list of labels as csv and change the color of each label for easier differentiation

Navigate to "Training Data" to start data labelling

Now you need to select the uploaded document for labelling.

This step will automatically take you to the document labeling page

Create a new labeling session

Click here to label documents to use to train your model

Click "wordpress-pdf-invoice-plugin-sample.pdf (page 1)"

wordpress-pdf-invoice-plugin-sample.pdf (page 1)

First step, make sure to select the curser for ease of navigation. Other tools are used for more advanced labelling, 

IMPORTANT
Once done labelling, click "Update" to save your annotations

To label text on the document, first click on the bounding box then click on its corresponding label on top of the document. This will assign a label to it automatically

Now do the same for all the labels that are available on this document.

For text that is split into multiple bounding boxes, just highlight all of the and then select the corresponding label.

You can check for your labels by looking at the right hand side widget and compare the labels that were assigned to the text

Training data annotation
2274
AD
Admin
#61
1/1
Skip
Update
Invoice#
1
PO#
2
Invoice Date
3
Total Amount
4

No Region selected

Regions
 137
Labels
Sorted by Date
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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
Invoice#
113
PO#
114
Invoice Date
115
Invoice Date
116
Invoice Date
117
Invoice Date
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
Total Amount
134
135
136
137
Relations (0)

No Relations added yet

COMING SOON!

Once you finish labelling enough documents you can navigate to "Cognitive Skill Training" from the left menu to train a model.

Deepopinion

studio.deepopinion.ai

This is a BETA tutorial for the Intelligent Document Processing workflow. In this tutorial we cover the process of uploading and labelling documents for information extraction.

DeepOpinion Document Information Extraction

Want to make guides like this in seconds? Yes, it's really that fast.