This page describes the data model used by the Document Insights application when integrated with Google Gemini, an OCR solution for parsing and reviewing Custom Forms. The model is designed to support template-based extraction, field mapping, and confidence-based review workflows.
🧩 Object Relationship Overview
-
Document Insights stores the OCR results of a submitted custom form.
-
Document Insight Template stores the mapping configuration used to extract fields from the custom form.
🧾 Object Schema
1. Document Insights
Stores the OCR result of a Custom Form parsed via Google Gemini, including extracted field statuses and confidence matrix.
|
Field |
Description |
|---|---|
|
Status |
Overall processing status (e.g., |
|
Status Description |
Additional detail for the status (e.g., "All fields matched", "3 fields missing"). |
|
Confidence Matrix (LWC) |
A Lightning Web Component that displays:
|
2. Document Insight Template
Defines the template structure and field mappings used to extract data from custom forms submitted via Google Gemini.
|
Field |
Description |
|---|---|
|
Intelligence Model Name |
Name of the custom form template (e.g., “Loan Application Form”, “Credit Report Summary”). |
|
Source Object |
API name of the Salesforce object where the extraction process will initiate. |
|
AI Intelligence Engine |
Google Gemini |
|
Target Record |
API name of the Salesforce object where the extracted fields will be mapped (The record can be updated or created based upon Source Object). |
|
Data Field Mapping (JSON) |
JSON structure defining how OCR fields map to Salesforce fields (key = OCR label, value = API field name). |
|
Prompt |
The prompt to be sent to Google Gemini AI engine to parse the required fields from the file uploaded. |
📊 Visual Overview
