IQ Bot standard forms
Standard forms are fixed format type documents that usually contain a large number of fields with very little or no variation in the overall layout, which can include both hand-written and digital content. IQ Bot enables you to extract data from standard forms.
Based on the associated role or permission defined by the Control Room administrator, you can then create a learning instance using a model for extracting data from standard forms.
Some examples of standard forms are W-4 (Employee Withholding Certificate), income tax form, sale deed and partnership agreement. Consider a scenario where you want to extract data from a large number of income tax forms. Create a model by name IT Forms and train this model to extract content from income tax forms. You can then create a learning instance using the pretrained IT Forms model in IQ Bot and use it to extract data, such as name, assessment year, age, account information, and so on from the income tax forms.
- Set up the required services and create a model.
- Set up and configure standard forms model in IQ Bot.
Contact your Customer Success Manager (CSM) or Partner Enablement Manager (PEM), who will help you set up and configure standard forms in IQ Bot.
- Manage learning instance.
- Log in to IQ Bot and create a learning instance for standard forms document type.
- Edit a learning instance if you want to change the name or description.
- Add validation rules for the learning instance based on standard forms bot.
- Move a standard forms learning instance across different IQ Bot environments.
- Extract data from documents in a production environment.
If you have a large set of documents, some of the them might not be processed and are stored in the Fail folder. For example, if you try to process more than 10000 documents, three might fail.
- Fix errors for the processed documents.
- Generate output.
Standard forms learning instance uses models from third-party vendors that are designed to extract form fields into CSV files. Table fields in the documents are extracted as separate CSV files and are not validated. Ensure that you manually review and correct errors in the CSV file.
- Use the learning instance details to review and analyze the output data.