Extract PDF Data for Data Mapping

How to Extract PDF Data

When you load a sample PDF to create a template and define data extraction rules, the PDF is displayed next to a schema pane. The schema pane displays a tree structure that represents how the data will be extracted. The MapForce PDF Extractor includes a powerful suggestion engine that automatically identifies common document elements and attempts to detect their structure.

For instance, the suggestion engine will identify tables that exist in the document, which you can then opt to extract automatically. A split operator in the schema pane helps you define how to correctly divide the table into separate rows. The suggestion engine can look for edges or lines to create the split, or split based on a fixed distance, for example, which you can preview in the PDF view pane. At the same time, the suggestion engine captures columns and header text. Clicking on any object in the schema tree highlights the corresponding structure and data capture rules as they apply in the PDF document view.

In large PDF documents with numerous tables, it may be helpful to search for text (such as a header) in the document and define rules for processing data relative to that text. For instance, when creating a template to extract data from yearly financial reports, you could search for “Expenditures” and process the table of figures following that text accordingly. Granular search options such as case sensitivity, format filtering (font, font weight, etc.), and whole or partial word searches, allow for precise targeting.

After the tabular data is extracted, you can adjust the extraction rules as necessary to exclude some fragments, adjust anchor assignments, define table boundaries, and so on. This can be accomplished using visual tools and helpful pull down menus. You can preview the results of data extraction in the output tree to check for accuracy.

Other document elements can be captured and added to your template manually. To define rules for extracting data manually, simply select an area in the PDF to extract by capturing it in a rectangle. Then, select Text Capture from the right-click context menu. PDF Extractor adds the capture as an element in the document tree, and you can drag and drop it to the desired position in the tree.

As you work, the MapForce PDF Extractor builds an XML document representing the structure of your PDF template with sample data from the working PDF document in the output window. This helps you understand and perfect the results of the extraction that will become a template to use in MapForce.