--- title: "Dataset Input" slug: "dataset-input" updated: 2025-05-21T11:11:47Z published: 2025-05-21T11:11:47Z canonical: "help.csgi.com/dataset-input" --- > ## Documentation Index > Fetch the complete documentation index at: https://help.csgi.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Dataset Input ### Dataset Input **Uploading and Processing Billing Data**: The Dataset Input feature allows businesses to upload billing statements and leverage AI to detect bill sections and tax-related items automatically. This streamlines content structuring, ensuring billing explanations are clear and accurate. ![](https://cdn.document360.io/c578bb14-e570-4b6e-93ed-f6c5b8a2d8a1/Images/Documentation/Upload Metadata.gif?sv=2026-02-06&spr=https&st=2026-10-26T13%3A18%3A52Z&se=2026-10-26T14%3A18%3A52Z&sr=c&sp=r&sig=zrC4KCpQORjXV7Clf4KxuMPOLd%2Be8uRFgY%2F2rbWbkJo%3D) #### Uploading Billing Data **Prerequisite** - Before using the Dataset Input feature, ensure users have a sample billing statement or a bill structure file ready for upload. The file should be well-formatted and include the necessary billing details to help the system accurately detect sections and tax-related items, reducing the need for manual corrections. ![](https://cdn.document360.io/c578bb14-e570-4b6e-93ed-f6c5b8a2d8a1/Images/Documentation/2025-02-27_14-09-54.png?sv=2026-02-06&spr=https&st=2026-10-26T13%3A18%3A52Z&se=2026-10-26T14%3A18%3A52Z&sr=c&sp=r&sig=zrC4KCpQORjXV7Clf4KxuMPOLd%2Be8uRFgY%2F2rbWbkJo%3D) - In the **Studio Module**, navigate to **Billing Content** and select the **Add** button located below **Dataset Input** to upload the file. - Upload sample invoice files or a bill structure file. - Use the drag-and-drop feature or browse the file via the file explorer. - Click **Upload and Process Now** for processing the dataset and section identification. > [!NOTE] > The supported file formats are PDF and XML, with a maximum file size of 500 megabytes (MB). #### Dataset Processing and Section Identification ![](https://cdn.document360.io/c578bb14-e570-4b6e-93ed-f6c5b8a2d8a1/Images/Documentation/2025-02-27_13-43-50.png?sv=2026-02-06&spr=https&st=2026-10-26T13%3A18%3A52Z&se=2026-10-26T14%3A18%3A52Z&sr=c&sp=r&sig=zrC4KCpQORjXV7Clf4KxuMPOLd%2Be8uRFgY%2F2rbWbkJo%3D) The detected sections are displayed in a side panel, where users can: - Navigate through the detected **Sections** and **Bill Items** using the left-hand side panel and review each one. - Identify new sections, which are clearly marked as “**New**” to differentiate them from pre-existing ones. - Manually review and delete any identified items that are incorrectly categorized as sections. - Click “**Add New**” to integrate valid sections into Bill Explainer or discard unnecessary ones. By leveraging AI, this feature helps businesses ensure billing statements are well-structured, accurate, and easy to understand reducing the need for manual data entry.