Skip to main content
Classification assigns each document to a class. You can treat each issuing company as a separate class and classify documents by company using a database of companies.

​
When to use company-based classification

Documents from the same company usually look alike and have the same types of fields in the same positions, which makes data extraction easier. The database of companies must be included in the FlexiCapture project. Populate it from the list of companies in your ERP system. FlexiCapture periodically syncs the database with the latest ERP data. If you don’t have a database yet, build it while capturing data, adding companies at the document verification stage. FlexiCapture searches for the required fields only on the first and last page of each document, where company information usually appears. Company-based classification has two advantages over image- and text-based classification:
  • No training set is required, so you save the time and effort of collecting sample images.
  • It supports up to 100,000 classes, far more than image- and text-based classification.
You can also use company-based classification for field extraction. Each company has its own section variant, for which you can train or create a separate FlexiLayout. Within a project, you can use both of the following at the same time:
  • A document-type classifier at the batch-type or project level.
  • A company-based classifier for documents of the same type, at the Document Definition level.
FlexiCapture first runs the document-type classifier to determine each document’s class, and then runs the company-based classifier for documents of the required class. Because the detected document variants uniquely identify the FlexiLayouts to use, field training runs independently for each company.

​
Configure company-based classification

Company-based classification runs within a Document Definition, that is, for documents of the same type. These documents share the same set of data fields to extract.
1

Open the data set setup

Right-click a document section and click Properties, or open the Document Definition editor and click Document Definition → Document Definition Properties. On the Data Sets tab, select a data set and click Set Up.
2

Enable the database of companies

Select Use database of companies. The required columns and their types are already specified in the data set (you cannot modify this table).
3

Map the data set to your database

To connect the data set to an ODBC-compatible database, map each data set field to its counterpart in the database. Map at least one field, such as the company name. For a field with no counterpart, specify None. For more information, see Using vendor and business unit databases.
4

Set normalization for name variants

To match company names that have more than one variant, specify a Normalization type in the Data Set Column Mapping dialog box. For more information, see Normalization of values in data sets.
Only certain fields are used to look for company information. These fields have a small lock icon next to them. Custom fields you add during setup are used only to display information.
If a company’s name is known in advance (for example, from the scanning operator’s name or the sender’s email address), you can specify the supplier and company subdivision explicitly before automatic detection. Set the fc_Predefined:PredefinedSectionVariantId document registration parameter to the Id of the appropriate Dataset entry. Automatic detection still runs for that entry, returning the explicitly specified company name and a confidence value for how well it matches the name extracted from the image. This method works only when a single section in a document has multiple variants.

​
Check and edit results

No training is required, because FlexiCapture looks companies up in a predefined list. Operators correct classification errors. When a document is attributed to the wrong company, the operator selects the correct company and saves it to the database, and FlexiCapture applies that correction to future classifications. To let operators correct errors, display classification results on the data form and add a button that starts the field look-up.
1

Create a service field

In the Document Definition Editor, click Create Field → Service Field. On the Data Source tab, select Flexible Section Variant ID from the Source list. Create any other service fields needed to identify the company, such as IBAN and VATID.
2

Create a Database Check rule

Right-click the group, click Properties, open the Rules tab, and click New Rule. Select Database Check and click OK. In Data source, select Data Sets, and then select the data set. In Field where to save record ID, select the service field from step 1. Click Add and specify the document and database fields. If their values differ, select Enter value from database → If values are different.
3

Add a look-up button to the data form

Right-click where you want the button and click Insert Button. On the Format tab, select the Database Check rule from step 2. On the Position tab, name the button.
Fields the classifier detects for company-based classification now have a region. The operator can click the button on the data form to open the Look up dialog box.

​
Improve company-based classification

​
Keywords and regular expressions

Specify keywords and regular expressions to improve company detection. Use strings that uniquely identify a company, such as data from VATID or IBAN fields.

​
Company records

For each company, you can specify multiple name variants and addresses. The administrator does this in the Document Definition editor, or a verification operator does it during verification. Only company records stored by FlexiCapture are modified. Even with ERP synchronization enabled, administrator and operator changes are not transferred to the external database. Operators can add and edit records only if the Document Definition allows it, and by default they cannot. To allow it:
1

Open the data set setup

In the Document Definition editor, click Document Definition → Document Definition Properties, open the Data Sets tab, select a data set, and click Set Up.
2

Enable operator permissions

Select Operators can add records and Operators can edit records. To prevent adding and editing, clear these options.