Guides

Intelligent Document Extraction

Extract key fields and table information from documents. Supported file formats: PDF, JPG, JPEG, PNG. Supported downloadable result types: JSON, TXT, Excel, CSV.

---

Home

From the homepage, you can directly access the Intelligent Document Extraction feature. The homepage is the central hub of ComlDP's navigation, enabling quick access to commonly used modules.

Single Document Extraction

Perform intelligent document extraction on a single document.
Select the file you want to extract. After successfully uploading, select your desired extraction mode, either Specified Field Extraction or Extract All Key Information in the Document. Click Extract Now and wait for a moment to see the Smart Document Extraction results.

Batch Document Extraction

Batch extract up to 30 documents.
Click the Bulk button above to switch to batch document extraction. Upload the file you want to extract, set the fixed fields to be extracted, click Extract Now and wait for a moment to see the Smart Document Extraction results.

• Text Field: Name of the key information field to be extracted
• Table Field: Name of the table header to be extracted

Introduction to Extracting JSON Fields

JSON Content Explanation

Parameter Name	Data Type	Description
code	String	Error code, "200" means success
message	String	Error message
data	Object	Return result
+details	Object	Key information extraction result
++Page-index	Object	Corresponding page number extraction result
+++key	String	Key information field extraction result, key:value
+++tables	Array	Key information table extraction results, tables:[ [table1], [table2] ] ]

JSON structure example:

{
    "code": "200",
    "msg": "success",
    "data": {
        "details": {
            "Page-1": {
                "Order Date": "xxx",
                "Order #": "xxx",
                "Quote#": "xxx",
                "Your estimated delivery date is": "xxx",
                "tables": null
            }
        }
    }
}