What this category's data looks like
In the vocational training subsidy application scenario, materials are submitted in batches by training institutions and assessment units at each node of class opening, teaching, and completion, including the training class opening filing form, daily attendance records, completion assessment form, subsidy application details, and receipts. The class opening filing form carries the institution, trade, class hours, and class opening date; the attendance records carry student names, attendance dates, and signatures; the completion assessment form carries assessment scores, assessment dates, and assessor signatures; the application details carry subsidy amounts and personnel lists. Materials are continuously updated by batch and class, and the same student's name, ID number, attendance days, and subsidy amount must remain consistent across multiple materials. When these materials enter the document parsing and chunking function, they are first parsed into retrievable data rows, and then used for subsequent retrieval and ledger generation according to index status, each serving the role of source evidence and verification basis.
Configuration and implementation checks
This step is executed by the parsing and training task (server-side task), outputting data rows and index status (parsed, indexing, indexed). The "version ledger and change records" are then completed by retrieval and subsequent processing. After parsing is completed, parsed chunks are immediately persisted as visible data rows and bound to a unique data identifier; the index transitions through three states, and data pending indexing does not participate in recall.
| Object | Check | Verification action |
|---|---|---|
| Training class opening filing form | The institution name, trade, class hours, class opening date, and responsible person's signature must be aligned item by item with the attendance records and completion assessment form | Verify the file records and data rows separately, confirm that the filing form fields are consistent with the attendance and assessment materials, and check whether the data row status is parsed, indexing, or indexed |
| Attendance records | Student names, attendance dates, cumulative class hours, and personal signatures must be consistent with the completion assessment form and application details | Verify attendance dates and signatures day by day, compare the same student's name and attendance days in the completion assessment form, and confirm that the data row status is consistent with the interface display |
| Completion assessment form | Assessment scores, assessment dates, assessor signatures, and the personnel list in the application details must be consistent | Verify the assessment date and assessor signature, compare the number and names of personnel in the application details, and check whether the data rows covered by the rebuild task contain only data with completed indexing |
| Application details and receipts | Subsidy amounts, personnel lists, and receipt acknowledgments must mutually corroborate with the filing, attendance, and assessment materials | Verify receipt signatures and amounts, confirm that the name, quantity, and date of the same record are consistent across all materials, and check the categorized display of processing modes related to chunkSize, chunkSplitter, and trainingType in the training progress dashboard |
Three common errors
- Symptom: The same student's name or attendance days do not match between the attendance records and the completion assessment form. Cause: The two materials were filled in by different handlers, and the same personnel identifier was not used for verification during entry.
- Symptom: The class opening filing form has been updated, but the application details still reference the old version. Cause: The version was not synchronized, and the old data rows are still in the completed indexing state and are recalled by retrieval.
- Symptom: The completion assessment form lacks an assessor signature or the receipt is missing. Cause: The acknowledgment step was not completed when the materials were submitted; although the parsed data rows have been persisted, the evidence is incomplete.
How to verify the setup
This feature acceptance: Upload one material, observe the data row status transitions and the coverage scope of the rebuild task, confirm that the status passes through parsed, indexing, and indexed in sequence, and that the rebuild covers only data rows with completed indexing, while the initial training covers data pending indexing and data being indexed.
Business test: Change one piece of content, observe whether the version number and references are updated synchronously, verify whether the name, quantity, and date of the same record are consistent across the filing, attendance, assessment, and application materials, and confirm that the number of references and the tokens of the reference text are recorded separately.
- Open the data rows to verify the status, confirm that it is one of parsed, indexing, or indexed and consistent with the interface display, and verify the identity identifiers of the file records and data rows separately.
- View the data rows covered by the rebuild task, confirm that only data with completed indexing is covered, and verify the current stage progress categorized by processing mode in the training progress dashboard.