Evaluate your use case
Content extraction models can be integrated into IQ Bot for extracting data from specific type of documents.
Before you proceed further with standard forms, evaluate your use case to determine if the content extraction from standard forms suits your requirement. Use the following questions as a guide to determine whether your use case needs Standard forms processing
- What are standard forms?
- Fixed format type documents that have a defined template, with very little or no variation in the overall layout. Standard forms are also known as structured documents, are generally characterised by high density of information, and are typically official or legal forms. The following table illustrates some examples of standard forms:
Tax form Medical form Legal form
- Does the layout of your documents vary across vendor?
- If yes, then this use case for content extraction using standard forms is not the best option.
- Are the positions of the various data elements (checkboxes, tables, and so on) fixed in your documents?
- If no, then these are not standard forms and this type of content extraction is not suitable.
- Is the document a dynamic form? For example, you have documents where the values depend on other data elements.
- If yes, then content from such documents cannot be processed using standard forms content extraction.
- Does the documents have repetitive sections?
- If yes, then this is not a standard forms use case. However, if there are a fixed number of such repetitive sections that do not vary between documents, then it can be processed as a standard forms.
- Does your documents contain tables that have a different number rows and columns?
- If yes, then these documents cannot be considered as standard form. The number of rows and columns have to be fixed for standard forms.
- Can document layouts that vary per year be considered as standard forms? For example, a W-4 document in 2020 and W-4 document in 2021?
- If the document layout does not vary within the same year or vary across organizations, then it can be considered as standard forms. If there are layout variations across years, then different standard form models must be used to process these documents.