Intelligent Structured Parsing
Last updated:2025-08-12 16:26:01
Intelligent Structured Parsing uses structured table rewrite models to automatically determine whether user-imported table files are in structured format. If your uploaded source file is not structured, the model will still automatically assess and possibly rewrite it into a structured table to enhance spreadsheet QA accuracy.
Structured and Unstructured Table Features
1. Structured table
Core features: have a strict two-dimensional relational model and directly correspond to the structure of relational databases.
Data features
Rows and columns are strictly aligned, with each cell corresponding to a unique row-column coordinate.
Single-level header (single row without merged cells), where column names have uniqueness and clear semantics.
No data nesting, each cell stores atomic data.
Support database import and export, support SQL queries.
2. Unstructured table
Core feature: Human-read-prioritized visual rendering method compromises machine readability
Structural interference feature:
Merge cells (cross-row/cross-column/multi-level merge)
Dynamic column structure (for example: column count varies by row position)
Repeat header per page (repeat column names per page)
How to Rewrite an Unstructured Table As a Structured Spreadsheet
Note:
After the user uploads documents, the system automatically determines whether the imported table file is in a structured format by calling the structured table rewrite model. No need to perform any additional configuration.
The structured table rewrite model has the following key capabilities:
1. Row list header identification: identify whether the table is a list header or row header.
2. Table name rewrite: generate a new table name by combining the file name, sheet name, and content.
3. Information extraction: extract remarks and supplementary information from non-trunk tables and place them in the description field.
4. Merge cell deconstruction: deconstruct merged cells and restore row and column alignment.
5. Single header rewrite: convert multi-row/multi-column headers into a single header format.
After intelligent rewriting by the model, unstructured tables can be intelligently converted to structured tables (but not supported in some scenarios).
After uploading a structured table file, you can click Parsing and splitting intervention in the document list. Upon entering the secondary page, view the structured display.
If you consider the structured display of the current model transcript has an error, you can re-edit the original spreadsheet and upload it; or target the spreadsheet to turn off the "retrieval enhancement" switch, then the spreadsheet will not be used for SQL query, but it will not impact the recall of semantic slices.
For structured table files, you can click "Advanced table settings" to add description or customize table header definition, enhancing table retrieval and QA accuracy.
If you want to return to edit the markdown result after table parsing and splitting, click switch to original slices display.
Are There Different Methods to Retrieve Structured Tables
In advanced settings of Application configuration-Knowledge, after enabling the "Excel retrieval enhancement" switch, SQL query for structured tables will be supported to enhance user query and calculation accuracy for such files based on natural language. If the switch is off, "Semantic Retrieval" or "Hybrid Retrieval" policy will be used by default to retrieve semantic slices for spreadsheet QA.
Was this page helpful?
You can also Contact Sales or Submit a Ticket for help.
Yes
No
Feedback